Factors Critical to the Success or Failure of Emerging Agricultural Cooperatives

Richard J. Sexton and Julie Iskow

Department of Agricultural and Resource Economics

University of California, Davis

Giannini Foundation Information Series No. 88-3

June 1988

The Giannini Foundation Information Series, initiated in 1963, publishes reports of timely interest for specific readerships. Reports are numbered serially within years. Single copies of this report may be ordered free of charge from Agriculture and Natural Resources Publications, 6701 San Pablo Ave., Oakland, CA 94608. Order by report number (Information Series 88-3).

Kirby Moulton is serving as Giannini editor. On the editorial board are Larry Karp, Quirino Paris, Michael Caputo, and Keith Knapp. All Giannini publications are peer reviewed. Carole Nuckton is technical editor.

Other publications of the Giannini Foundation and all current publications of foundation members are listed in the Giannini Reporter issued annually in the summer. Copies of the Reporter are also available from the ARN publications office.

The recently established University of California Center for Cooperative~ provided the funding for this publication in order to further understanding of cooperatives.

The UC Center for Cooperatives is a product of 1987 state legislation (AB 1597). The Center's purpose is to enhance the development of both agricultural and consumer cooperatives in California by strengthening research, teaching, and public outreach services pertaining to cooperatives. The university-wide Center will be located at UC Davis and will be fully operational by July 1, 1989.

FACTORS CRITICAL TO THE SUCCESS OR FAILURE OF EMERGING AGRICULTURAL COOPERATIVES

the authors are:

Richard J, Sexton As.sistant Professor

Julie Iskow Postgraduate Research Economist

Department of Agricultural Economics University of California, Davis

ACKNOWLEDGEMENTS

This report is an outcome of a cooperative work agreement between the University of California, Davis and the Agricultural Cooperative Service, U.S. Department of Agriculture. Funding was provided by USDA, ACS, under the terms of research agree-ment No. 58-3J31-5-0014. Many people contributed to the completion of the project and this report in particular. The authors arc particularly grateful to John Haas and William Seymour at ACS for their patience and encouragement and to the many land-

grant university economists, Bank for Cooperatives officials, and state agricultural council leaders who helped in locating recently-formed cooperatives. We are especially indebted to the many leaders of new cooperatives who shared their experiences and view-points with us and to Christina Fitz Gibbon for timely processing of this TepOrt.

The views expressed here are the authors' and do not necessarily represent those of the USDA or any other agency.

Funding for this project was provided by the U.S. Department ofAgriculture, Agricultural Cooperative Service under the terms of research agreement No.ss..3)31·S.00!4. The views expres.sed here are the authorn' and do not necessarily represent the viewpoint of the U.S. Department of Agricultwe or any other agency.

i

TABLE OF CONTENTS

PREFACE ........................... iv

IIlGHLIGHTS .............................. iv

INTRODUCTION ...................................................... . . ........... 1

PART1, THE ECONOMIC ROLE OF COOPERATIVES IN MARKET-ORIENTED ECONOl\flES .......2

I. The Economic Essence of a Cooperative... . ......................................... . . ................... 2 II. When Is Vertical Integration Desirable? ... . ......................................... . .......... 4

A. Margin reduction................................... . ......................................... . . ................... 5 B. Market power avoidance ............................... . ............................................................... . . ................... 7 C. Using marketing cooperatives to influence consumer price ........................................ . . ................. 12 D. Risk reduction through cooperatives .................................................................... . . ....... 13 E. Cooperation in markets where no alternative market opportunity exists.................. . . ....... 15

III. Summary........ ......................................................................... . ......................................... . . ....... 18

PART 2. THE ORGANIZATIONAL, FINANCIAL, AND OPERATIONAL KEYS TO DEVELOPING A SUCCESSFUL COOPERATIVE................................... . .................................................... 19

I. Organizational Keys to Success ................................... .. . .......................................... 19 A. Membership........ . ...................................................... . . ........................................ 19

1. Developing the initial coalition ................................ . . ......................................... 19 2. Building the membership .............................................. . . ......................................... 19 3. Long-term membership policy ................................ . . ................................................... 21 4. Nonmember business................................................. . . ......................................... 22

B. Optimal vertical integTation in a cooperative ................. . . ....................................... 23 C. Decision making ............................................................... . . .......................... 23

1. Voting.................. . .................................................. . . .......................... 23 2. Management. ....................................... . . ......................... 24

II. Financial Keys to Success .... . ................................................. . . ........................ 25 A. Generating initial equity............................................................... . . ................ 25 B. On-going equity acquisition and redemption ....................................... . . .......... 26 C. Debt finance .............................................................................................. . . ...... 27 D. Use of gTants and technical assistance ................................................... . ...... 29

III. Operational Keys to Success. ................................................................. . ...... 30 A. Pricing policy ................... ................................................................... ....30 B. Output policy................... . .......................................................................... . .....31 C. Evaluating a cooperative's performance ...................................................................................... . .....32

IV. Summary.......................................................... . ............................................................................. . ....33

ii

TABLE OF CONTENTS

PART 3. AN EXPLORATORY ANALYSIS OF RECENTLY-FORMED AMERICAN AGRICULTIJRAL COOPERATIVES ...................................................................................................................................................34

I. Identifying Newly-Formed American Agricultural Cooperatives ............................................................ 34 II. The Survey Instrument ...................................................................................................................................... 34 III. TheMainResults ............................................................................................................................................... 35

A. Characteristics of the survey respondents ...............................................................................................35 B. Economic factors ········································-························································ ....................................... 35 C. Membership factors···································-························· ...................................................................... 37 D. Financing ..................................................................................................................................................... 38 E. Decision niaking and management .......................................................................................................... 39

IV. Statistical Analysis of the Determinants of Success in Emerging Agricultural Cooperatives .................41 A. The success ratings .....................................•................................................................................................. 41 B. A statistical model of the determinants of sua::ess ..................................................................................41

1. Constructing a measure of success ..................................................................................................... 41 2. Variable selection .................................................................................................................................. 42 3. Results ...................................................................................................................................................42

V. Sununary...............................................................................................................................................................45

CONCLUSIONS ....................................................................................................................................................... 46

SURVEY FORM SENT TO ACTIVE MARKETING COOPERATIVES .......................................................47

iii

PREFACE

This report studies the determinants of success or failure among recently formed American agricultural cooperatives. I tis intended to assist fann leaders and their advisors in making decisions concerning new cooperative ventures.

Focus is given to economic environments in which a cooperative may improve farmers' welfare and lo the organizational, financial,and operational requisites to cslablishing a successful cooperative. It is assumed that readers have some basic understanding of the key principals and concepts of cooperatives.

HIGHLIGHTS

Cooperatives play an important role in America's agricultural economy. However, the fail-ure rate among new cooperatives is often high and, moreover, situations wherecooperativescould bene-fit fanners may sometimes go unrecognized. This studywas undertaken to provide a guide to the types ofeconomic conditions in which cooperatives can be beneficial and to set forth the key organizational, financial, and operational features to developing a successful cooperative. In this sense the report is intended to act as a blueprint to guide farm leaders and their advisors in making decisions concerning new cooperatives.

In Part 1 of the report the key economic functions of cooperatives are described and the market condi-tions amenable to a cooperative's presence are set forth. Cooperatives are an organization through which farmers may address market failure by jointly vertically integrating themselves into the market chain. As a consequence of market failure, farmers may pay too much for farm supplies or receive too little for fann product sales. They may also be ex-posed to excessive price and income risk orbe unable tobuyorsell all of the products they wish. This report

shows how to recognize market conditions under which one or more of these problems may occur and how the problems may be overcome through a coop-erative.

Part 2 of the report focuses on theorganiz.ational, financial, and operational keys to successfullydevel-oping a cooperative, given that one is needed based on the market conditions as described in Part 1. Among the factors discussed in Part 2, particular stress is given to insufficient membership and busi-ness volume and to a poor initial equity base as critical problems for new cooperatives. Several sug-gestions to help overcome these and other problems are presented.

Part 3 of the report analyzes the experiences of 61 recently-organized American agricultural coop-eratives. These cooperatives present a wide diversity of outcomes ranging from major success to minor success to failure. Statistical analysis isolates several factors critical to the successful cooperatives' per-formance including the initial involvement of a large number of members, growth in the membership, use of full-time professional management, and accep-tance of nonmember business.

,,

INTRODUCTION

This report is about the causes of success or failure of emerging agricultural cooperatives in the United States and itsprotectorates. Be<:ause coopera-tives are voluntary organizations, they will succeed only if they provide benefits to their members in excess of what is available elsewhere. An often-used phrase is that cooperatives "must be born of neoes-sity.'' Accordingly, a major focus of this report is the benefits cooperatives can possibly provide in a mar-ket-oriented economy. Although membership in a cooperative may provide intangible benefits such as satisfaction from participation in a democratic or-ganization, our focus will beon theeconomic benefits of cooperation.

Even cooperatives that are born ofnecessity may fail if they lack sufficient membership and volume, are improperly financed, or are poorly managed. Consequently a second focus of this report is the organizational, financial, and operational requisites to the successful development of a new cooperative. We should stress at the outset, however, that this report is not intended to be a "how to" guide on the mechanics of starting a cooperative.'

Cooperatives have played a fundamental role in the development of America's agriculture. That role has increased in importance through the twentieth century to where about 30 percent of farm products are now marketed through cooperatives at the first-handler or farm-gate level and 27 percent of farm supplies are purchased from cooperatives.

However, cooperatives' overall share of the agri-cultural economy has not increased in the 1980s. Moreover, decline has been noted in recent years in the number of new agricultura1 cooperatives being formed, and the failure rate among those that have been organiz.ed has been high. This report is, there-fore, timely in that cooperatives have been a tradi-tional means of self help for farmers, particularly during times of e<:onomic hardshlp like many farm-ers have been experiencing during the 1980s.

This study has three main components. Part 1 reports on the results of recent research into the possible economic benefits of cooperatives in a modem, market-oriented economy, while Part 2 focuses on the organizational, financial and opera-tional keys to development of a successful coopera-tive. Taken as a whole, Parts 1 and 2 provide a loose ''blueprint" for success in new modem-day agricul-tural cooperatives.

Part 3 of the study reports on results of a survey ofagricultural cooperatives formed within thelast 15 or so years in the U.S. and its protectorates. Leaders of these cooperatives were asked to complete a ques-tionnaire designed to discern the economic factors motivating the cooperative's inception and the or-ganizational, financial, and operational keys to its subsequent success or failure.

'A pub!ical:lon devoted to this task is Gene Ingalsbe and James L. Goff, Hmv wS"'rl o Coopmilitie, Cooperative Information Report No. 7, U.S. Department of Agriculture, Agricultural Cooperative Service, 19SS.

1

PARTl. THE ECONOMIC ROLE OF COOPERATIVES IN MARKET-ORIENTED ECONOMIES

In this part of the report, we first identify the key economic functions perfonned by cooperatives in our economy, and then proceed to enumerate and discuss the specific benefits that cooperatives may be able to provide to members. In each instance we first describe the potential benefit and then indicate how to identify the specific market conditions where the benefit can be realized.

I. The Economic Essence of a Cooperative.

Cooperatives are oftentimes misunderstood. Jn part, confusion has been caused by people's desire to impart soda! or political connotations to coopera· lives. However, the ovCITiding significance of coop-eratives, especially to American fanners struggling to succeed in today's economy, is as an economic organization.

The basic characteristic that distinguishes coop-eratives from other businesses is that they are owned, controlled by, and intended to benefit the people they serve-the members-rather than outside in· vestors. However, the feature of member control does not distinguish cooperatives from many other diverse organizations including labor unions, coun-try clubs, and governmental units.

The additional concept needed to set coopera-tives apart from other organizations and to under-stand their role in the economy is vertical integration. In tum, to understand vertical integration it helps to think of a production process flowing like a river beginning with the mining of raw materials and ending finally at the consumer. The outputs of each successive stage in the production flow are, in tum, inputs into the next stage and so on.

If a business operating at one stage in the process decides to extend its operations into additional stages of the process, we say the business has verti-cally integrated. Downs Ire.am or forward integration is when a firm moves into production stages closer to consumers, like the oil refiner who owns a pipeline distribution network and possibly a chain of retail gas stations. Upstream or backward integration is when a business suppliesits own productive inputs such as a refiner who owns crude oil reserves and a distribu-tion network to transport oil to its refineries.

Figure 1 schematically depicts a typical pattern of the production flow for agricultural commodities and lists examples of products produced at the vari-ous stages. Just as the oil refiner in our previous

illustration sometimes found it desirable to integrate its operation upstream and/or downstream into the product flow, the agricultural producers depicted in the middle of Figure 1 may have a similar incentive to vertically integrate. We will soon enumerate and

Figure 1. A Typical Market Flow for Agricultural Production

RAW MATERIAL PllOOIJCT!ON

~-<rUd•<>il·­RAW MATERIAL PllOCCESSINC

·~,,,finod<>il

'"­FARM INl'l!T MANUFACTIJRJNC

t..ctor.. impl<""'1" g>.•,<>iL lubric•n" ,., ono, !..nd tool•

FARM R!."TAlL SUPPLY [ J FARM PllOOUCT!ON

liv..IDd

·~ ~·£..,;., ond vq;mbl..

FOOD PROCCSSJNC 1...h•nd;it-=-""""n"l<I"'' butru,m....

mill..:! S"'"'• =eols, i..,.d, "'""""' '"""" •nd dri<dITuit.•nd v•g"'•bl..

[ FOOD WHOLESALING

[Foooo.rrA!UNG _d FOOD SERVICE ounirrs ,...1ounnt&

hoop;W .~.

O)NSUMERS J discuss these incentives, but for now accept that farmers may on the upstream side wish to become involved in producing and retailing the fuel, fertil-izer, chemicals and other inputs tha I they use in their farm operations. On the downstream side they may have incentive to market their own production in-cluding possible ownership of processing facilities and transportation equipment.

However, most agricultural producers would encounter a fundamental problem in attempting to vertically integrate: building a processing plant for a single producer's output would almost never be efficient. That is, the minimum size of operation

2

needed to efficiently operate, for example, a process-ing plant almost certainly will exceed the scale of the fanning operation by a vastmagnltude. Moreover, a single grower would usually not have the financial capacity to set up an efficient plant.

The solution to this dilemma is for many produc-ers to integrate jointly, to band together to build or purchase the processing plant. This action of joint vertical integration is the economic essence of a coop-erative. Stated in terms of a definition, agric11/tural cooperation represents coordination of producers to achieve mutual vertical integration. That is, by banding together in a cooperative, fanners who each have incentive to vertically integrate can jointly overcome the vast scale discrepancies that normally will exist between the farm sector and the upstream or down-stream industries. When farmers integrate down-stream, they fonn a marketing coaperative, while joint upstream integration results in a purchasing or supply cooperative.

Table 1 illustrates this point using data from the late 1970s for a sample of food processing industries. It compares the minimum level of sales needed to attain efficient scale in the industry with the average level of sales for fanners selling to the industry. Most noteworthy is the final colunm of the table which indicates how many average-size farms are needed to supply the minimum efficient-size plant.2 For example, the output of over 2,000 average-size live-stock producers would be needed to supply the minimum efficient-size meat packing plant. In other words, farmers in these industries would not on their own be able to efficiently integrate into the process-ing of their raw agricultural products. But if they were to band together through a cooperative in approximately the numbers indicated on the table, they could jointly run an efficient processing plant.

Table I. Farm Size and Efficient Procesing Plant Size Comparisons for Selected Industries

1977 value of shipments for

minimum 1977 value of efficient plant 1978 average Number of shipments in scale in sales per farm average fanns

Processing millions of thousands of farm in thousands of perminlmum industry dollars dollars industry dollars efficient plant

Livestock except Meat packing 31,130 102,700 dairy and poultry 32.2 2,153

Creamery butter 1,110 14,300 Dairy 70.3 153

Ch- 5,246 33,000 Dairy 70.3 325

Canned fruits 6,321 457,000 Vegetables and melons 87.3 1,635 and vegetables 6,321 457,000 Fruit and tree nuts 49.9 2,862

Aour 4,569 33,800 Wheat 30.2 616

Milled rice 1,242 64,000 Rice 129.7 333

Raw cane sugar 708 10,000 Sugar 187.9 32

Soybean oil mills 6,117 102,700 Soybeans 32.0 3,721

Sources: Processing industry data are adapted from Conner, J. M., et al., The Food Manufacturing Industries, Lexington, Mass: D. C. Heath & Co., 1985; Fann industry data are from the 1978 Agricultural Census.

'These figures were obtained by adjusting the sales from the average farm by a ronversion /actor to reflect the farm-processing price spread and then dividing sales for the minimwn efficient-size plant by the adjusted sales/or the average farm in that industry.

3

II. When Is Vertical Integration Desirable?

It is now time to operationalize the notion that cooperatives must be "born of necessity." The key task lies in discerning when vertical integration is a good idea.

To begin, economic exchange can always be handled in one of two basic ways: at arm's length through a market or internally through a firm. Verti-cal integration is desirable when a transaction is more efficiently handled through the internal work-ings of a firm than through the market.

However,undercertain conditions, markets will always be the most effective way to transact busi-ness. Markets that satisfy these conditions are called competitive markets. The key characteristics of com-petitive markets are:

1. They contain a large number of both buyers and sellers.

2. Firms are free to enter or leave these markets without penalty.

3. All firms within a market produce the same product, i.e., there is no product differentia-tion.

4. Traders in the market possess all informa-tion. For example, everyone knows the price(s) being offered or charged.

When farmers buy or sell in markets that tend to satisfy these four characteristics, they can be assured of receiving the highest price possible for the prod-ucts they se\l and paying the lowest price possible for the farm supplies they buy. Vertical integration cannot improve upon the performance of these "perfect'' markets and farmers have no incentive to form cooperatives in these instances.

However, most actual markets do not satisfy all of the competitive market characteristics. In these cases markets may not perform perfectly, that is, there maybe market failure. When markets fail, it may be beneficial for farmers to vertically integrate, that is, to bypass the market and conduct transactions through a cooperative.

In discussing particular types of market failure and how they may be remedied through a cooperative, we will focus mainly on marketing cooperatives but will also develop key inferences for supply cooperatives. A useful concept to help analyze marketing cooperatives and understand the detennination of farm prices is the marketing margin. The margin essentially accounts for all of the costs incurred by the marketing sector in bringing the raw farm product to the consumer. The margin will typically include costs of performing the following marketing functions:

1. assembly of the raw product from the farms, 2. processing.. 3. distribution and wholesaling, and 4. retailing. Alternatively, the margin can be broken down

into costs for the inputs such as labor, energy, capital, and materials used by the marketing sector. For a given agricultural product, we can denote the mar-gin as Mand the retail price for the finished product asP,_ Given Mand P,, the maximum value, Pt, for the farm price, P1, is

P ~=(P,-M)/K (1)1where K is a conversion factor indicating the number of units of the farm commodity needed to produce a unit of the retail commodity. For example, ii takes roughly two pounds of beef on the hoof (K = 2.0) to produce a pound of beef at retail.

If the industries that assemble, process, distrib-ute, and retail the fann product are perfectly com-petitive as described above, the firms in these indus-tries must operate efficiently to survive. That is, they must perform their functions at the lowest cosl pos-sible. Therefore the margin, M, will be as low as possible.

These Silme competitive pressures will force P1 up 10 the maximum given by equation (1). At-tempts to pay farmers less than this price will be thwarted because competition will bid the price up to what the product is worth. It bears repeating that under this scenario a marketing cooperative would not be able lo help farmers. However, when the conditions of perfect oompetition are not all met in the downstream industries, our simple equation linking farmand retail prices can help pinpoint how cooperatives may make fanners better off.

Three possible ways to raise the farm price are apparent;

1. If a cooperative can market the farm product at a lower cost than the existing noncoopera-tive firms, the margin, M, separating the farm and retail prices can be reduced.

2. If marketing sector finns have market power over farmers, theymaybeableto force farm-ers to accept a smaller payment than what the product is worth, i.e., a price lower than Pt defined in equation (1). A cooperative may counterbalance this market power.

3. If coaperative marketing can increase P,, that is, if the price consumers pay for products produced from the fann conunodity can be increased, a higher farm price will result.

Any action that raises the farm price will raise farm income and make the farmer better off. How-ever, before looking in more detail at the three ways

4

cooperation may raise farm prices, we need to intro-duce two additional factors. The first is risk.

Everyone knows fanning is a risky business. Prices fluctuate from year to year, and production levels can be similarly volatile. These factors often combine to make fannincome very unstable. Income risk is usually personally dissatisfying and also cre­ates management problems.

Thus, if marketing through a cooperative can reduce this risk, we have an additional benefit of integration through a marketing cooperative. We will elaborate on this point shortly. The second additional factor to consider is when no private handler is available to purchase the farm product. For example, the present buyer may go out of busi-ness. Under what conditions, if any, will it then be wise for the fanners to organize a cooperative t-0 carry on the marketing activity? We shall pursue this point in more detail, but, first let's retum and discuss the three price-related benefits that may emerge from cooperation.

A. Margin reduction.

There are two possible ways to lower the margin through cooperative marketing:

1. The cooperative firm may face lower prices for some of the inputs used in marketing, •nd

2. The cooperative may market the product more efficiently than is presently being done.

It is unlikely that a cooperative will be able to buy labor, energy or other materials more cheaply than otherbusineS&!s, but it may be able to extract savings on the cost of capital due to either the manner in which income accrued by co-operatives is taxed and/ or to possible advantages of cooperatives in securing cheaperdebtcapital. We shall discuss thetaxelement here and defer the question of debt finance to Part 2, Section !IC.

The value of a capital asset, say a processing plant, is based on the income stream net of taxes generated by the asset. Denote the expected stream of gross {pre-tax) income flowing from a processing plant over N years as Y1, Y2,... ,YN, where the sub-scripts denote successive years into the future. Be­cause a dollar of income received today is always worth more than a dollar received some time in the future, we must progressively discount projected earnings from the plant for succeeding years into the

future. This is doneby dividingincome earned in any year, t, by the amount 1 plus the appropriate interest rate all raised to the power oft.

Finally to obtain the value of the capital asset, the gross income must be converted to net income by multiplying it by (1 - T1) where T1is the tax rate for a firm of type i, for example, a cooperative or a nonco-operative corporation. Drawing these elements to-gether into a single formula, the value, V~of the plant to a finn of type i is

Y, Y, YN ]Vi= --+--+ ... + N (1-T;) (2).2[ (l+r) {l+r) (l+r)

The key point from equation (2) is that if one type of firm faces a lower tax rate than other firms in the industry, (1-T,l isa bigger number for that firm and the plant is more valuable to it than to the otherfirms. In effect, thecost of capital is lessforthe low tax finn.

A cooperative will usually pay less tax on a given amount of net income than will an ordinary corpora-tion. Thereasonis that the Tax Code allows coopera-tives to pass net income through to their members without paying tax upon it. The income is taxable to the members whether it is received in cash or certifi-cates of allocation that will be converted to cash by the cooperative at sometime in the future.

This income pass-through feature is the same treatment afforded income earned within the vari-ous stages of a vertically integrated corporation. There are two reasons it often results in a lower tax rate on net income received by a cooperative com-pared to its noncooperative counterparts:

1. The cooperative's earnings escape the double taxation that occurs when an ordi-nary corporation's income is taxable to the corporation and thenagain to the sharehold-ers when they receive it as dividends.

2. The tax rate paid on income received by cooperatives depends on the farmer/mem-bers' tax rates. This rate on average may be lower than the corporate rate paid by the noncooperative corporation.3

Although this tax advantage may not itself pro-vide reason enough to form a cooperative, it can help when other reasons are also present. In particular, if the tax advantage is large, equation (2) suggests that it makes sense for the cooperators to buy marketing facilities from a noncooperative firm in the industry rather than build facilities from scratch.

'A brief set of references on the topic of the taxation of cooperatives is: Morrison Ne<ly and James Board.a, Logo! P~ of Former Cwpmdiws, lnfonnation 100, U.S. Department of Agriculture, Farmer C.OOperative Seivice, May 1976; Lee F. Schrader and Ray A. Goldberg. FMrmer c.wp.m.tiM; ond F«lm1/ l11m1tU Tuts, Ballinger Publishing Co., Cambridge, Mass., 1975; and Richard/. Sexton and Terri A. Sexton, "Taxing Co-ops," Parts I and II, Owict1s, Volume 1, 2nd and 3rd Quarkrs, 1986.

5

The ~nd way the margin may be reduced is when a cooperative can handle the marketing func-tions more efficiently (at lower cost) than ordinary corporations. This possibility revolves back to our earlier discussion of the relative efficiencies of han-dling exchange through markets versus the internal workings of a vertically integyated firm.

The three main advantages to internalizing transactions through vertical integration include:4

1. lntemalization creates a common incentive among parties, whereas participants to market exchange usually have opposing interests, i.e., the buyer wants to buy low and the seller wants to sell high.

2. Disputes within an organization can be re-solved quickly through internal control, while disputes between independent parties often involve costly litigation.

3. Information usually flows more freely within an organization than across markets.

These ad vantages to vertical integration become most important when a large proportion of farmers' assets are sunk. A sunk asset is one whose cost cannot be recovered by resale within a given time period. An asset is partially sunk if only a portion of the cost can be recovered. For example, a custom-built milking parlor is likely to be a sunk asset because it cannot be resold quickly or without considerable loss. On the other hand, hay purchased to feed a dairy herd is not a sunk asset since it can normally be resold quickly with little loss.

Farmers with a high proportion of sunk assets are vulnerable to opportunistic behavior on the part of their trading partners. In other words, farmers whose assets are sunk are "stuck" in the sense of lacking alternative opportunities. Trading partners may try to take advantage of this situation. For ex-ample, growers who produce highly perishable commodities such as vegetables are vulnerable to opportunistic price cutting by buyers because the harvested crop is often a sunk asset-its perishability gives the grower few resale opportunities.

Similarly on the input supply side farmers who need immediate supplies of inputs such as liquid fertilizers, chemicals, and petroleum are potentially vulnerable to hold-ups by sellers who may try to take advantage of the situation by extracting higher prices.

In principle, buying and selling contracts can be carefully and expensively written to limit the scope

for opportunistic behavior, but contracts need not be honored. The result then is often costly litigation.

These possibilities highlight the incentive com-patibility and conflict resolution advantages of inter-nal versus market organization.

We must be careful, however, before attributing these advantages to cooperatives. The reason is that, although cooperatives do accomplish vertical inte-gration for their members, they don't really internal-ize transactions within a single firm. On the contrary, farmers' need to integrate jointly results in exchange through a coopcrati ve often quite closely resembling market exchange.

Thus, cooperatives do not replace market ex-change. Rather, they hannonize exchange. Jn parti=-lar, the cooperative and its members usually have common incentives-the farmer wishes to sell at the highest price possible and the member-owned mar-keting cooperative wishes to pay its members the highest price possible subject to covering its costs. Similarly the farmer wants to buy supplies as cheaply as possible and the farm supply co-op wants to provide themas cheaply as possible subject to the same breakeven proviso.

If conducting exchange through cooperatives can reduce costs due to protracted bargaining, writ-ing and interpreting long-term contracts, litigation, and inefficient information exchange, the margin, which reflects the costs of marketing the farm prod­uct, can be reduced and in tum, the farm-level price raised. The main efficiency-enhancing feature of cooperatives to accomplish this margin reduction is, as noted, the harmonization of trade between a coop-erative and its members relative to trade between fanners and independent buyers or sellers.

How to recognize when margins might be reduced by cooperation. It is one thing to agree that cooperatives may make trade more efficient and quite another thing to identify the specific markets wherein effi-ciency could be enhanced through a cooperative.

Tobegin addressing this important question, it is interesting to note that informal, legally unenforce-able contractual relations dominate in the business world and that resort to legal sanctions is rare.> The reason is that most businesses have on-going, mutu-ally profitable relations with their trading partners. A business attempting to behave opportunistically would see its trading relationships severed to its ultimate detriment. In other words, a business' repu-tation or its goodwill are important assets that will

'Oliver Williamson has been the most articulate exponent of the effidency-enhan<.ing features ofintrafirm versus market excl>ange. His recent book, f.amcmic Org.,,i%ation, New York UniverSity Press, 1986, draws together much of his work on the topic. 'See Stewart Ma,,,.uley, "Non-Conb'adual Relahons in Business' A Preliminary Study,~ American Sociolagy Rei>iew, 28, February 1963.

6

rapidly be diminished by opportunistic exploitation of trading partners.

However, when goodwill becomes an unimpor-tant consideration in a firm's decision making that we have cause to worry about opportunistic behav-ior, and harmonizing trade through a cooperative may result in transactions cost savings and margin reduction. Goodwill considerations are less likely to enforce scrupulous behavior on farmers' trading partners in the following situations:

1. A large proportion of farmers' assets are sunk as the term was defined earlier.

The key question for a fanner to ask in respect to his/her trading partners is "what loss would I incur if a partner refused to perform the agreed upon transaction?" For example, for the fresh vegetable grower the answer could range from "nothing," if an alternativebuyer could be solicited with a phone call, to "the entire crop," if no alternative buyer were available on short notice. This first factor indicates farmers' vulnerability to opportunism and to the transactions costs incurred to avoid it. The next two factors relate to a business's incentive to behave opportunistically. Opportunism is more likely in

2. industries in long-term decline. Finns will not expect to remain in these industries for long, and, hence, goodwill considerations are not important;

3. times of economic hardship. Short-run sur-vival considerations are the dominant con-sideration during these times. The long-term consequences of unscrupulous behavior arc given little consideration.

In sum, when farmers' assets are sunk and when long-run reputation effects for trading partners be­come unimportant, farmers arc potentially vulner-able to opportunistic price cutting from buyers and price gouging from sellers. The reason we list this behavior under margin reduction is that opportunis-tic tendencies can in principle be controlled through contracting and litigation. However, these inefficien-cies of the market mechanism raise the margin and hold open the possibility that farm product prices may be raised and input prices reduced by harmo-nizing exchange through a cooperative.

We move on now to discuss the second item on our list of ways farm prices can be raised through a cooperative.

B. Market power avoidance.

The opportunistic behavior we have just dis-cussed results in trading partners attempting to exer-cise short-term market power over farmers. An alter-native, long-term market power may be present

when farmers perennially have only one or a few firms buying farm production.and selling farm sup-plies. Economists use the term monopsony to de-scribe a market with only one buyer and oligopsony to characterize the case of only a few buyers. In these situations, farmers may be paid less than the fair price, Pt = (P, - M)/K, for their product. The reason is simple: with only one or a few buyers, the competi-tive pressures that normally cause price to be bid up to Pi" are either absent entirely or are not very strong.

The same principles are at work in the markets for farm supplies, but the terms are somewhat differ-ent. Monopoly refers to a market with a single seller; oligopoly describes markets with only a few sellers. In these markets, sellers will probably try to charge more for farm supplies than it costs to provide them, and competitive forces are often not strong enough to prevent this type of overpricing.

Aside from charging high {paying low) prices, another feature of monopoly {monopsony) power is likely to be charging (paying) different prices to different farmers for no good, i.e., cost-justified, rea-son. This type of conduct is called price discrimination. Price discrimination can be the outcome of playing farmers off against one another and attempting to discern each's minimum selling price for farm pro-duction or maximum buying price for supplies.

This type of pricing behavior can persist because competitive forces are usually weak in monopoly/ monopsony markets. lf markets were competitive, attempts at price discrimination would fail because competition among buyers or sellers would bid the price to a uniform "competitive" level. Several re-spondents to our survey indicated the presence of this type of discriminatory pricing behavior in their buying or selling markets.

Cooperation is a way to integrate around the market power. Simply put, farmers can organize a cooperative to market their product and no longer have to deal with the monopsony or oligopsony firms. The cooperative will pay its members the largest net price possible subject to covering its mar-keting costs. Similarly, the purchasing cooperative integrates farmers around monopoly or oligopoly power and supplies its members farm inputs as cheaply as possible subject to covering costs.

Cooperatives' role in circumventing market power is the most familiar justification for coopera-tives in a market-oriented economy and has played a key role in the development of many American coop-era lives. Nonetheless, several aspects of cooperation in market power environments need to be carefully weighed by groups considering establishing a coop-erative.

7

The first concern is the costs involved in market-ing the fann product or supplying the fann inputs. A great number of cost studies havebeen done over the years and regardless of theindustry analyzed the studies almost invariably conclude that the curve relating per-unit costs to the level of output is roughly ''L" shaped as depicted by the curves in Figure 2. In other words per-unit costs tend to de-cline, possibly rapidly, over small levels of output and then flatten out thereafter. Only rarely does evidence of "U" shaped per-unit costs surface wherein per-unit costs rise for large-scale finns as depicted in Figure 3.

The concern is that a cooperative formed by fanners to integrate around buyer or seller market power may not achleve sufficient business volume to move to the flat portion of the per-unit cost curve. If thls happens, the marketing cooperative's margin will be larger than for the other finns, and the price it will be able to pay its members may end up being less than the other marketing finns' offers. Similarly, a supply cooperative that did not achieve sufficient volume could end up charging more than competitors.

This concern is greatest when the cooperative enters an industry with a high level of fixed costs for capital equipment and relatively low incremental operating costs. These capital intensive industries have steeply sloped per-unit cost curves resembling curve A in Figure 2 with a corresponding large cost penalty for failing to attaln the minimum efficient scale (indicated by Q"A on Figure 2). Curve Bin Figure 2 represents, on the other hand, an industry where the cost penalty for an inefficient scale is quite low. This flat per-unit cost curve would be reflective of an industry, for example, fresh fruit and vegetable marketing. with relatively low capital intensity.

A second consideration in respect to cooperation in monopsony or oligoprony markets is whether vertical integration is actually necessary at all. Fann-Figurc 2, Some Typical lrshaped Per -unit Cost Curves

" 1i'\:....__________ CurveB

.._-!--------- CurwA

1 OUTPUT

ers who face market power may be able to achieve improved tenns of trade merely by organizing to-gether to bargain with the buyers. An organization of this type is called a bargaining cooperative. It does little or no actual handling of the fann product, but at-tempts through collective bargaining to improve the price fanners receive from the private handlers.

The success of a bargaining cooperative hinges on its ability to organize and control enough raw product volume to force price concessions by the marketing finns. A critical factor is that members must commit to marketing through the association so that it has actual control of the raw product.

The main leverages a bargaining association can exercise are to:

1. play the marketing finns off against one another, causing them to bid up price, and/or

2. threaten to entirely withhold the product from the private handlers by fonning a marketing cooperative to directly process and sell the product. In effect, a bargaining association is one step from becoming a marketing cooperative. Its members have coordinated their selling activity, but they have not yet vertically extended into the market chain.

It is probably the best of all possible worlds if fanners' price enhancement goals can be attained through a bargaining cooperative. The financial commitment is less than for a marketing cooperative, and concerns about achieving production efficiency in marketing are not relevant.

The third concern about countering market power with a oooperati vecenterson possible barriers to entry into the market. In particular, if the existing buyersare exploiting fanners by paying less than the fair price defined in equation (1), it would seem to present an opportunity for a new firm (one that is not a cooperative) to enter the market and offer to pay

Figure 3. An Example of a U-shaped Per-unit Cost Curve

o· OUTPUT

8

farmers ahigherprice and thereby capture a share of the market. As long as the entering firm's price did not exceed P1•, it would make at least ample profits. Similarly in farm input supply markets an entering firm could undercut the monopolist'sor oligopolist' s price, capture a share of the market, and still make an profit.

If no such entry occurs, several explanations are possible:

1. The opportunity has gone unrecognized, 2. The existing firms in the industry have

erected barriers to prevent entry, or 3. No entry opportunity actually exists, i.e., no

market power exploitation is actually taking place.

The third reason is always a possibility and high-lights the importance of potential cooperators being as certain as possible that there is a price-enhancing (price cutting) role to be played before committing to a marketing (purchasing) cooperative. The same can be said of the first reason in which case developing a cooperative is an appropriate response to the market's failure to recognize the profitable entry opportunity.

Of most concern is the second reason pertaining to barriers to entry. If existing firms have succeeded in keeping out other marketing firms, is there any reason to believe entry by a cooperative will fare differently? Certain structural barriers to entry can-not be overcome. These include instances when an existing firm

1. holds patents necessary to a production process,

2. controls the entire supply of a material needed in the production process, or

3. hasbeen granted an exclusive license or fran-chise to operate in the industry.

In these cases, even though no entry is possible, farmers may still achieve price enhancement through a bargainingcooperativc if they follow theguidelines noted previously.

A second type of entry barrier is strategic in nature. If businesses incur sunk costs to enter a mar-ket, they will by definition be unable to recover these costs if they are forced to leave the market. The necessity of sinking costs, for example, in the form of nonrccoverablecapital invesbnents, is a risk of enter-ing new markets. Existing firms in an industry can exploit this risk to keep out entrants. In particular, tlueats by incumbent firms of price wars or other forms of ruinous competition can deter entrants whenever sunk costs are important. The reason is that sunk costs pTevent a new finn from costlessly

leaving the market if it finds it cannot make a profit given the incumbent firms' behavior.

Thus, in markets with high sunk entry costs and concerns about predatorybehavior by existing finns, competitive entry may be forestalled and existing finns may continue to exercise market power. Can a cooperative do any better at breaking the incumbent firms' lock on the market? Recent research suggests the answermay beyes, due to a numberofconcephlal differencesbetweena coopcrativeand a noncoopera-tive entrant.6 For one thing, a marketing (purchasing) cooperative comprised of the farmers who sell to (buy from) the incumbent firm(s) is a very dangerous entrant from an incumbent's viewpoint. If most or all of the producersdecide to market through a coopera-tive, the supply of product for the other marketing firms driesup. In input supply markets, demand facing the noncooperative sellers would dry up if mostof the farmers began patronizinga supply coop-erative. In contrast, if the entrant is an ordinary corporation, it and the incumbent firm will usually achieve some mutually-profitable market-sharing arrangement.

Another factor is that a cooperative is compara-tively invulnerable to predatory behavior by the incumbent(s}. Predatory pricing to drive out a coop-erative would involve paying farmers11Wre than their product was worth in the marketing context and charging a price less than cost for inputs. While this strategy can quickly inflict losses upon a noncoopera-tive entrant, it plays right into the hands of the c0-0p members who can sell at inflated prices or buy sup-plies at cut-rate prices.

The upshot from these considerations is that it will often be in existing firm(s) best interests to deter entry by a cooperative, especially if its members would jointly comprise a large share of the relevant market. As just noted, threats of predation are inef-fective against the cooperative, so deterrence can be accomplished only by committing to raise prices paid (lower prices charged) to fanners suffidently so that they no longer have incentive to develop amarkefing (purchas­ing) cooperative. These price improvements must remain in force as long as c0-0p entry is a threat.

For example, if we denote a new cooperative's marketing costs as M', the price the cooperative would be able to offer members is

P; = (P,- M') / K. The incumbent firms would have to commit to a

price offer of at least P/ to prevent the cooperative from developing.

'See RlchardJ. Se~ron and Terri A. Sexton, "Cooperatives as Entrants," RJi:nd foumal ofE<>lnomics, 18, Winier 1987, pp. 581-595.

9

The key implications from this analysis for farmer groups are the following:

1. A threat of vertical integration may be as good as vertical integration itself in accom-plishing the farmers' price enhancement goals. The way to realistically pose this threat is through a bargaining cooperative. Farmers considering developing a market-ing cooperative should first coalesce into a bargaining association.

2. If the incumbent firm(s) cannot commit topay satisfactory prices over time, that is, if opportunistic behavior is a strong possibility, the best bet for farmers is to evolve directly into marketing their production through a cooperative. A bargaining association will probably be ineffective.

How to recognize when buyer market power is a problem. We know that fannersare not receiving a fair price if they get less than Pt= (P, - M)/K. We can observe the retail price, P,, that consumers pay for the finished farm products. However, it is very difficult to compute all the costs that go into the margin, M, as the product flows through the several marketing stages depicted in Figure 1.

One possible way to directly judge the perform-ance of marketing firms in a given area is to compare the price they pay with the prices being paid for similar farm products elsewhere. Even this mode of comparison is more difficult than it might initially seem. Farm prices for similarproductsmight differ in different areas for a number of reasons.

1. Regional demand differences may make retail prices and, hence, farm prices higher in some areas than in others.

2. Marketing costs will not be the same across regions for a number of reasons: a. Transportation costs will be less and

farm prices consequently higher for producing regions located near major consuming centers.

b. Costs for inputs used in marketing will not be the same across regions. For ex-ample, labor is cheaper in the South than in other parts of the U.S.

c. Processing plants may be more efficient in major producing areas due to efficien-cies of large scale operations as depicted in Figure2.

3. There may be subtle quality differences in the raw products produced in different areas with the higher quality products naturally extracting a price premium.

These factors all indicate the great difficulty in-volved in directly judging marketing firms' perform-ance. The alternative that economists have fre-quently employed is to focus on the observable struc-tural characteristics of anindustry and inferprobablc industry conduct and performance from the structural characteristics. Often times elements of industry conduct will also be observable and used jointly with structural characteristics to infer price perfonnance.

An industry's key struchlral characteristics re-late back to the four features of perfectly competitive markets set forth at the beginning of this section. Two are particularly crucial in our analysis of product marketing:

1. The number, relative size and location of buyers.

If only one company buys in an area, the market isa monopsony. If only a few buy, it is an oligopsony. Relative size of the buyers is as important as their numbers. Even ifseveral buyers are available, if a few control a large share of the market, they may succeed in controlling pricing of the farm product to fanners' detriment.

Geographical location of buyers is also a crucial structural factor for many agricultural products. Farmers bear the cost of transporting their raw prod-ucts from the farm to processing facilities. Often these costs arc very high, particularly for bulky or perish-able products. Thus, even though several firms may be willing to buy a farmer's production, transporta-tion costs, if processing facili lies are not proximate to the farm production, will diminish the net farm price. Therefore, buyers located large distances from the fanning region do not provide much protection from possible monopsonistic exploitation by the one or two local marketing firms.'

2. The ease of entry into and exit from the marketing industry.

Even when only one or a few firms are available to buy the farmers' product, opportunities for price exploitation are minimized if there are no impedi-ments to entry into and exit from the market. In this case, attempts to pay less than the fair price would be foiled by entrants who could purchase the product at a higher price, perform whatever marketing services are needed, sell the finished product and still make a profit.

'Exampl"" of organizing a cooperative primarily to redu"" marl<eting tranoportation "°""' are provided in O. M. Simon, W. R. Garland, and Jan Halkett, EsWJlishing • Coltcn·Ginning Cooperotiw in IM Southusl, U.S. Department of Agrirolture, Agr!roltural Cooperative Servi..,.,, ACS Research Report 7, May 1981; and Jerry G. West and Lionel Williamson, "Can a CooperatlveSu<tted. Serving Small Family Fann•?" Fa._r Coopomzti~, 44, September \978, pp. 26-27.

10

Entry and exit are not free if entrants must incur sunk (nonrecoverable) costs. Capital-intensive in-dustries will almost always involve a significant portion of sunk costs because specialized equipment can usually not be resold quickly without loss.sOther costs besides capital costs may also be sunk. For example, the entrepreneurial labor involved in start-ing a business is a sunk cost.

The need to sink costs can deter entrants who are unsure of a market's future profitability. In tum, the entry mechanism can no longer be guaranteed to protect fanners from price exploitation.

In addition to these structural characteristics, the main type of market conduct that needs to concern fanners is coordinated pricing among marketing firms (i.e., cartel behavior) intended to offer fanners less than the fair price defined in equation (1). Overt collusion among buying finns is not necessary to achieve coordinated pricing. Rather, coordination may evolve informally over time with, for example, one finn acting as a price leader and others quickly matching the leader's price offer.

Detecting coordinated pricing may be difficult. The absolute level of the price being offered will usually not provide much evidence because of the many factors just noted that can cause price level differences between regions.

Either buyer or seller cartels are usually prone to at least temporary instability because individual firms can make money by cheating on a pricing agTeement. Two pieces of evidence reflect this insta-bility and in tum, the possible presence of a buyer cartel:

1. Periodic receipt by sellers of "under the table" offers to buy at above the announ.::ed price.

2. Price wars, that is, sudden and often large increases in the price being paid followed by an equally sudden decrease. Such behavior reflects the breakdown and subsequent re-structuring of a price agreement.

To summarize, if farmers have only one or a tew selling opportunities within their geographic region and if barriers to entry into the marketing industry arc high, the prospects are good that the farmers will be subject to monopsony or oligopsony exploi talion. Further evidence is provided by signs of coordinated pricing among buyers.

Farmers' best strategy in these cases is to coalesce into a bargaining cooperative and attempt to elicit price improvement through thatorganlzation. If this step is unsuccessful, the next step of integrating into a marketing cooperative should be considered. We shall discuss the keyelements in developing these or-ganizations in Part 2 after completing our discussion of the prospective benefits from cooperation.

How to recogniu when seller market power is a problem. Most of the rules just discussed for buyer market power apply also to market power for the supply of farm inputs, so this discussion can be brief. Any time farmers pay more for farm supplies than the cost of providing them including a fair rate of profit, they are paying too much.

Once again, though, it may be very difficult to judge directly if prices arc too high based on this standard. Regional price comparisons for the same product arc not entirely accurate because of possible differences in the costs of providing services. None-theless, these direct price comparisons probably work better in farm input markets than in the farm output markets.

Persistently higher prices in one area compared to others that arc not explainable through transporta-tion costs, different state sales or excise taxes proba-bly reflect market power. If any large, regional sup-ply cooperatives operate in the general area (e.g., Cenex or Farmland Industries), the simplest way to attack this market power may be to convince the supply cooperative to extend operations into the affected area.

If direct price comparisons cannot be made, the key structural and conduct characteristics described in the previous section can be used to infer perform-ance.

An additional complicating factor in judging the performance of input supply markets is product dif-ferentiation and advertising. Many sellers of farm implements, chemicals, seeds, etc. make special claims for their products and heavily advertise these claims. Concerns are twofold:

1. Does the product differentiation reflect genuine quality improvements which merit paying a price premium or docs it reflect hype or needless frills?

2. Are expenditures on nonpricc competition including advertising and product differen-tiation needlessly raising prices?

'Not all capital 00515 are 5Uilk 00515, For example, the bOO prepara~on and track for a railroad ore sunk rosts, but costs for the train ltself are not 5Uilk because the train can easily be shifted to operate on another track. A tecl>nlcal di=lon on this. subject is provided by WilliamJ. Baumol, John Panzor, and Robert Willig, Contt.<lllblt Markets Qnd !he Thtoiycf !ndW.<liy $lrucl~..,, ACS Research Report 7, Haroourt, Brace, Jovanovich, New York, 1982, especially Oiapter 10.

11

Quality differences may be surprisingly easy to investigate merely by evaluating the ingredient list-ing in fertilizer, feeds, and chemicals. Usually prod-uct differentiation and advertising go hand in hand with a market having comparatively few sellers, i.e., an oligopoly. To the extent that these features intro--duce additional needless selling costs, they provide an additional opportunity (along with combating price enhancement) whereby a supply cooperative may be able to cut purchasing costs for its members.

C. Using marketing cooperatives to influence consumer prices.

If farmers can increase the prices for their prod-ucts at retail, naturally farm prices will also increase. Two possible avenues exist to accomplish this goal:

1. The cooperative may be able to restrict the flow of farm product to the market.

2. The cooperative may be able to improve quality of the finished products.

We shall discuss each possibility in tum. Over-supply relative to demand lies at the heart of Ameri-can agriculture's financial dilemma. If farmers could jointly agree to restrict production, they could raise prices, reduce costs, and increase profits. However, farmers individually have no incentive to abide by output restriction agreements, and thus, proposals to control output have usually been destined to fail. For a cooperative to have success in restricting the flow of product to the market and, hence, raising price, the following factors must be present.

1. Thecooperativemustcontrola large share of the relevant supply. If a market is national or international in scope, or if production is scattered amongmany thousands of prod uc-ers, such control is not possible. However, for local or regional markets such as fresh fruits and vegetables, the possibilities are better!

2. The cooperative must have a way to keep excess production off the market. Volume controlcan be accomplished by a. closing membership and/or b. restricting members' deliveries.

However, simply refusing to process the excess production through the cooperative will have no effect on market price if alternative sales outlets arc available. The excess produclian must be kept entirely out of the market.

We should note that cooperatives that succeed in raising price in this manner may run afoul of the Capper-Volstead Act of 1922. This legislation gives

farmers the right to organize into cooperatives, but it also authorizes the Secretary of Agriculture to inves-tigate instances of undue price enhancement by co-operatives. That this authority has never been exer-cised is probably evidence of the limited success cooperatives have had at monopoly price enhance-ment.

The second way to raise retail and, thus, farm prices through cooperatives is by improving quality assurance. There are two reasons why a cooperative may be effective in this regard:

1. Production and marketing may be better coordinated through a cooperative than through ordinary market channelsdueto the improved flow of infonnation, characteristic of a vertically integrated enterprise.

Thus, the marketing cooperative may be able to successfully coordinate quality specifications with its members, set planting and harvest times to maxi-mize quality and so forth.

2. Private handlers of farm products at times will have incentive to shirk on quality.

Maintaining quality is costly, and private handlers for whom reputation is not a consideration can make money by diminishing quality and deceiving con-sumers. Of course, this strategy only works in the short term, and consumers will react by cutting pur-chases of the item. The quality-cutting handler can leave the industry counting its short-run, ill-gotten gain, but farmers are left to bear the brunt of dis-gruntled consumers and, consequently, reduced de-mand and lower fann prices.

How to recognize when a marketing cooperative can raise retail prices. The number of situations when cooperatives can exercise market control to raise retail prices is probably limited. The main exceptions to this rule are when markets are local or regional in nature and when production is concentrated within the hands of a relatively few producers. Markets tend to be local or regional for perishable and costly-to-transport products. Fresh fruits and vegetables, bread, and fluid milk tend to be local markets, for example, but improvements in transportation tech-nology are increasing the geographic scope of these markets over time. In addition, new entry into pro-duction of the product must be limited. (An example oflirni led en try would be tree crop production where there is usually a four to seven year lag from planting to initial harvest.)

Attempts by farm producers to regulate the supply of their product will usually be enhanced by the presence of a federal or state marketing order.

'Even though fresh fruits and vegetables may be shipped in from distant regions, transportation co>ts arc high. lbis factor gives local producers an element of joint market power.

12

Marketing orders give the force of law to volume control plans. 'This, in tum, overcomes the usual problem of producers tending to not abide by the controls.

Quality control problems can usually be ob-served, and consumercomplaintsdo filter back to the producer, so there is no mysteryinvolved here. Prob-lems are most likely when there are many competing marketing firms, with frequent entry into and exit from the market. In these cases, handlers arc hard-pressed to make money and will be tempted to cut costs by cutting quality. Transience of the marketing finns creates a time horiz.on problem. Destroying consumer's confidence does not hurt these finns if they planned to leave the industry anyway. It docs hurt producers for whom fanning usually represents a lifetime commitment.

It has been demonstrated over and over that consumers will pay premium prices not only for quality but for the assurance of qwility. Thus, if the structural symptoms of quality cuttingare present or if poor or inconsistent quality of the product deliv-ered to consumers is observed, cooperative market-ing may be the answer. For example, Sunkist Grow-ers began in the early 1900s to address precisely this problem.10

Having completed our discussion of ways coop-eration may raise the overall level of a farm product price, we tum now to consider cooperation's role in reducing the variability of farm prices and income.

D. Risk reduction through cooperatives.

One of the reasons often given to explain vertical integration by businesses is the need to attain more assured access to the upstream or downstream mar-kets. The same sort of risk-reducing behavior has also been attributed to agricultural cooperatives. Recent examples are the fuel and fertilizer shortages of the early- and mid-197Qs.11

The basic point to risk is this: Let a farmer's income stream over a number of years, say N, be denoted by I,, 12,...,IN. Chances are the income stream will trend upward over time. We could subtract out the trend component of the income stream and call thedetrended series I,_•, I,•,...,IN•. The mean or aver-age of the series is

J-[It +I2"+...+IN"l/- /N

The farmer faces significant income risk if the year-to-year earnings levels fluctuate widely about the average. A common measure of this risk is the variance of the income stream computed as

(I 1•-I• )2t{1 t-1·)2

+ .•• +(IN" -I·) (3).Var(!•)= .

N The higher the variance, the riskier is the income

stream. Risk is usually bad for two distinct sets of reasons:

1. Most people simply prefer a stable income stream to one that fluctuates widely. These people are said to be risk averse.

2. Risk makes it difficult to efficiently plan and manage the farm operation. Credit becomes more difficult to obtain and it is harder to coordinate long-term investment decisions.

Thus, if cooperatives can reduce exposure to risk, they help farmers at both a personal and profes-sional level. Webegin bysettingforth twobasic types of market risk that arise in different contexts.

1. In some markets prices are inflexible and markets do not clear. (li prices are free to move up and down, markets should always clear in the sense of equating demand and supply.)

In these cases farmers may face sales rationing in the downstream market and rationing of purchases in the upstream input markets. (By rationing we mean being unable to buy or sell all that is desired at the price.)

2. Markets that do have flexible prices will clear so rationing is not a problem but wide price swings are possible with the effect being to increase fanners' exposure to risk.

The question for farmers who recognize one or both of the above conditions in their markets is how can a cooperative help? We consider first the case of marketswith inflexible prices and possiblerationing. Prices may be fixed, for example, due to government regulation or long-term contTacts. Even if prices are not absolutely fixed, they may be inflexible because businesses consider it imprudent to make frequent price changes. On the selling side, a marketing coop-erative can deal with the risk of rationing in one of two possible ways.

1. The cooperative can have a closed membership with the membership stTategically chosen so

"See C, H. Kirkman, TM S1<11kis1 Ad"""' I""'' US. Departme-nt of Agriculture, Farmer Cooperative Servi<:<!, R:S Information No. 94,1975. ns,,e, for example, Dcnald L Vogelsang, GJopnllli?I< Hirm Fertilizer C<ist>, U.S. Department of Agriculture, Eronrnnics Sta!isli"." ~nd Cooperatives Service, FCS Research Report No. 8, 1979 and Warren J. Mather and Homer/. Preston, Coopn"li"" Btn•filS "'1.d Llm1/.2-ii<ms, US. Department of Agriculture, Economics Statistics and Cooperatives Service, Cooperative Information Report No. 1, Section 3' 1980.

13

that the cooperative will have a high proba-bility of being able to market all of its mem-bers' production.

In years when additional raw product is needed, the cooperative can attempt to procure it from non-members. (The issues of open versus closed member-ship and whether to accept nonmember business are discussed in Part 2 of this report.)

2. The cooperative can have an open membership and agree to provide a "home" for members production even though some production may have to be destroyed or diverted.

In the first instance the cooperative shifts risk from members to nonmembers. In the second case it pools the risk and the costs of sales rationing. In other words, the open membership cooperative does not avoid the consequences of sales rationing. but it spreads them in an equitable fashion among the members so as to minimize their effect. A noncoop-erative handler, alternatively, might buy all the pro-duction from some farmers and nothing from others.

The same types of considerations are at work with upstream integration and a purchasing coop-erative. For example, during the fertilizer and petro-leum shortages of the 1970s, cooperatives limited sales to members and kept comparatively low prices. The first element of th.is strategy reflects the mem-ber/nonmember dichotomy noted above, while the pricing policy reflects the cooperative's goal of pro-viding service as cheaply as possible.

Turning now to analyze markets with flexible prices, consider the elements that detennine the farmers net income, I. Revenues are just the price, P, times the quantity sold, q. Costs are closely related to theamount produced and for simplicity we can write them as a constant function, C, of the amount pro-duced. Thus, we have

l=(P-C)q. Income is subject to risk because each of the three

elements that detennine it are subject to random fluctuations: P varies due to changes in output mar-ket conditions, C goes up or down based on condi-tions in the input markets, while q depends upon growing conditions and other natural phenomena.

Cooperatives can possibly help farmers cope with this inherent risk in either of two ways:

I. If the cooperative controls a significant share of the relevant market and the commodity is storeable, it can reduce price fluctuations by maintaining buffer stocks of the raw com-modity.

2. Withoutcontrolling a significant share of the market, cooperatives may still help farmers cope with risk through diversification.

The idea of a buffer stock is illustrated in Figure4. The cooperativeeffectively chops the peaks and valleys off from the cyclical production path by withholding commodity from the market in high supply years and releasing stored commodity onto the market in tight supply years.

Figure 4, The Effect of a Buffer Stock Program

"buffered" output path

TIME

Ideally, the buffer stock results in a uniform flow of product on to the market and hence, stable prices and incomes. However it is hard to make buffer stocks work. The U.S. government's experience with buffer stocks has not been particularly successful, and it's naturally harder for a single company to make them work.

Control of a large share of themarket is necessary if the stocks withheld from or added to the market are to have a significant impact on price. A "free-rider problem" also hinders these programs. For example, producers outside the cooperative have no incentive to withhold production during high supply years. Instead they "free ride" on the enhanced prices brought about by the cooperatives' supply manage-ment. In tum, this creates discontent among the cooperatives' members.. To limit free riding the coop-erative must control a large share of the product.

One of the few cooperatives to successfully per-form this buffering role is the California Almond Growers Exchange (CAGE). (CAGE has recently changed its name to Blue Diamond Growers.) CAGE' s experience highlights the importance of the points made here. Almond production is highly unstable, but the commodity is quite easily stored. Moreover, CAGE marketsover60percentof the U.S.

14

almond crop and 40 percent of the world's supply, among the largest market shares of any U.S. coopera-tive.

Most cooperatives, especially new ones, cannot expect to play this buffering role. An alternative, risk-management role is possible through pooling.

Pooling occurs when a cooperative markets several different products (or supplies several differ-ent inputs) and lumps all proceeds into one or a few "pools." Farmers then receive allocations from the pools in proportion to their patronage with the coop-erative. For example, a cooperative may market sev-eral different fruit and vegetablecommodities. It will usually give growers a partial payment (a so-called "established value") at the time of harvest. Subse-quent payments come from the pools. This feature results, for example, in peach growers sharing in income from tomato production and vice versa. By lumping the returns from several commodities to-gether in this fashion, the effect may be to diversify risks and stabilize growers' income streams.

In fact, agricultural producers themselves are of-ten diversified, producing and selling multiple commodities. However, modem capital-intensive farming systems tend to encourage specialization. Thus, as producers lose the risk diversifying effect of producing several commodities on the farm, it may make sense to regain diversification through a coop-erati ve.

Problems are twofold, however. First, one grower sharing in another's revenue introduces a cross-subsidy which, in tum, distorts market signals and leads to overproduction of some products and underproductionofothers. This situation will lead to memberdiscontent unless the cross-subsidies cancel over time. Second, multiproduct pooling may actu-ally hinder a natural tendency for revenue streams to stabilize. That is, years with poor production tend to cause high prices and hence, offsetting revenue ef-fects. With pooling. however, the growers who have the poor supply do not get the full benefit of the high price. They instead share it in the pool with produc-ers of other commodities.

How to recognize when a cooperative can reduce income risk. The risk generated from participating in a market is an easily recognized element of the market's performance. Most agricultural industries are subject to frequent and wide prices swings, so the question is not how to detect risk but how to recog-nize when a cooperative can mitigate the risk or its effects.

Privately-held buffer stocks, as noted, will have a stabilizing effect on price only when the coopera-tive controls a significant share of the product as in

the case of CAGE. Of course, the product must be storable. Buffering is not a realistic short-term goal for newly developed cooperatives.

The success of pooling hinges on finding com-modities with inversely correlated income streams. Two relevant statistical measures to judge this fea-ture for any two commodities are the covariance and the correlation coefficient. Denote detrended income streams a_!> before by I1•,12•,••• , IN• and the mean of the series by J•. Also let the subscripts i and j denote any two different commodities. Then we have the follow-ing fonnulae: Covariance:

Cov (It,! t> =[(I Ii* -ftl(I it-Itl+ ... +

c1 Nr-lt><rNt-Itl~.

Correlation coefficient:

Cov(It,Ijl Cor(It,J·•) = --,~s~cc.c"~'CC1 ,JVar(It>Var (ljl

where the Var{!*) formula is provided inequation (3). A negative value for either measure means the in-come fluctuations for the two commodities have tended to offset over the N years of data. Thus, these commodities are possible candidates for pooling. The correlation coefficient has the advantage ofrang-ing from -1.0 to +1.0, with -1.0 representing the unlikely case of perfectly offsetting fluctuations, 0 representing no correlation, and positive values indicating fluctuations that move together. Thus, large negative values for the correlation coefficient indicate especially good commodity candidates for pooling. Information needed to compute these statis-tics, such as a measure by crop of income per acre, should be available from farm advisors, county ex-tension agents, or land-grant university extension economists.

Pooling itself is probably not a worthwhile rea-son to start a cooperative, but if a cooperative is being considered for other reasons, the risk pooling poten-tial, if any, should be considered.

E. Cooperation in markets where no alternative market opportunity exists.

Up until now we have been weighing the ad van-tages of cooperative marketing or purchasing versus dealing with pri vale, noncooperative handlers oper-ating in the same markets. However, fanners have often faced and continue to face situations where no private handler is willing to serve a market. Can co-operation possibly be a good idea in these situations?

15

On the surface the answer would seem to be "no." If no for-profit company can earn a satisfactory return byserving the market, how then can a cooperativedo better?

In fact, there are three reasons that suggest a cooperative might be successful where noncoopera-tive counterparts have not. Thus producers faced with loss of markets due to exit by all for-profit handlers ought to at least consider cooperatively acquiring one or more of the facilities. The reasons are as follows:

1. As much of our previous discussion has indicated, a cooperative may be able to operate more efficiently than the unsuccess-ful noncooperative handlers. In the case of product marketing, an efficient cooperative maybeable to operate with a smaller margin than the noncooperatives were able to achieve.

2. The fanner-members of a cooperative arc probably willing to accept a lower return on investment than are the owners of noncoop-erative finns.

This statement reflects the common perception that most fanners get intrinsic satisfaction from fanning and, thus, will and do accept lower returns on their investment than are typically earned in other industries. As such, a rate of return on a processing plant that may be unacceptable to nonfarm investors may be acceptable to the fanners especially if preser-vation of their livelihood is at stake.

3. The harmonization of exchange afforded by cooperation enables flexible pricing meth-ods to be instituted that can extract value from product marketing or input purchas-ing unattainable to noncooperative firms.

Our third point is crucial, and to ful!y illustrate it requires introduction of a few additional economic concepts. Focusing on the case of a possible market-ing cooperative, denote quantities of the farm prod-uct being supplied with the letterQ. Farmers' supply curve, denoted as S(Q), indicates the minimum price farmers as a group need to provide various levels of output. The supply curve is determined by the (incre-mental) costs of producing various amounts of the product. Supply curves normally slope upward, i.e., to supply ever greater amounts, farmers must be compensated with a higher price.

Next, if we take the price at which the finished product can be sold and subtract off the margin, we obtain a curve that measures the maximum price a handler could pay for the raw product while cover-ing all marketing costs including an acceptable re-turn on investment. This curve is often called the

net-average-revenue-product curve for the raw product and is denoted by NARP(Q).

Finally, the net-marginal-revenue-product curve measures how much each incremental unit of the farm product is worth. The N?\-fRP(Q) is the difference between the final selling price and the incremental (not average) cost of marketing it.

All three curves are drawn in Figure 5 with the general shape they are usually considered to have during any short-run time horizon. S(Q), recall, indi-cates the minimum price farmers need to receive to cover their variable production costs and NARP(Q) indicates the maximum price a processor can pay and still cover its costs. The two curves do not inter-sect as the diagram is drawn. In other words, wehave a market that cannot be profitably served. Figure 5. Optimal Behavior for a Marketing Cooperative When No Single Price Will Cover Costs

S(Q)

NARP (0)

NMRP(Q)

a• RATE OF OUTPUT

However, the NM:RP(Q) curve, which says how much each incremental unit of the raw product is worth, does lie above the supply curve for some levels of production. In fact, if Q• units of the raw product are produced (the intersection point ofS(Q) and NMRP(Q), the total revenues in excess of vari-able production and marketing costs from produc-ing and selling the product are graphically depicted by the large dotted region minus the small cross-hatched area.

The problem for the noncoopcralive marketing firm.constrained to deal with farmers at arm's length is Iha t no single price enables it to extract any of these profits. However, a cooperative is ideally suited to practice flexible pricing. In the present example, it should pay producers a price of P, per unit and in so

16

doing extract the maximum total profits available from producing and marketing the product. To cover its marketing costs, the cooperative must recover from members the amount (P1 - P,)Q", which can be done, for example, by charging an annual member-ship fee.

Although we shall not illustraleit, the same prin-ciple holds for a farm input purchasing market. The key is a cooperative's ability to choose the "best" outputlevel, implement flexible pricing and coveran operating deficit with fixed charges to the member-ship. If the situation is as depicted in Figure 5, farm-ers will be able to pay these charges and still cam a satisfactory return on their total investment.

Haw to recognize when cooperation can be successful in markets that ordinary corporations will rwt serve. Farmers may face either of two alternative scenarios in this context. A market may have never been served or firms are leaving a market that previously had been served. Examples of the fonner situation have been provision through cooperatives of rural tele-phone service and electricity. Modem examples might include provision of services such asvideotexl (e.g., cable TV, remote transmission of auctions), integrated pest management and computer soft-ware. In terms of product marketing, instances abound wherein producers would like to locally market fresh fruits, vegetables, home baked prod-ucts, crafts,etc. but have no outlets. Other examples are specialty products, e.g., angora wool, that appear to have a number of willing producers but no market outlet.

In these instances no direct comparisons can be made between the past performance of a noncoop-erativc handler and the prospective perfonnance of a cooperative. Therefore, infonnation needed to make a wise decision may be difficult to attain.

Consider, for example,a possible cooperative lo mar­ket angora wool. A number ofpeople appear willing to raise the rabbits, and clothing manufacturers do use the processed wool (presently importing most of their needs). The gap in the market is that no domes-tic company presently buys and processes the wool. The potential rabbit producers and wool processing cooperative owners probably have a good idea of the rabbit production costs. They may be able to find out somethingaboutprocessing costs if such facilities are located elsewhere. Hardest to discern is probably demand for the processed product.

A great deal of uncertainty unavoidably clouds decision making in these situations. Considerable market research is called for but is expensive to

undertake. A technique often used by the USDA Agricultural Cooperative Service in assisting poten-tial cooperatives is to conduct surveys to determine both potential demand and supply for the services of the cooperative.

The decision over whether to cooperatively ac-quire and operate facilities of a defunct or soon-to-be defunct handler is more structured than navigating the uncharted waters of developing an entirely new venture. Potential cooperators should consider each of their three possible advantages over a noncoop-erative business.

In judging whether a cooperative may operate more efficiently than its prederessor(s), decision makers should review the previous discussion of ways in which cooperation may reduce the market-ing margin.

They can also attempt to discern the recent re-turn on equity in the noncooperative business. Re-turn on equity (ROE) is the net profits of the business divided by the owners' equity. This infonnation should be available from the financial statements of a publicly-held corporation. For a privately-held firm the information should be provided as part of any purchase negotiations. The key point is that the corporation's minimum-acceptable return on invest-ment may be greater than the potential cooperators require.

For example, over the past 50 years, real (infla-tion-adjusted) return on corporate equity has been about 7 percent. On the other hand the return on equity in U.S. agriculture in 1985 was -12.9 percent including capital losses, the sixth consecutive year real returns have been negative.12

The final consideration to weigh is whether a cooperative's built-in pricing advantage will enable it to extract sufficient revenue to coveritscosts, while leaving enough income to make the farming opera-tions profitable. One key to answering this question is to discern if there are differences among farmers in their willingness to pay for a farm input or service or in the reservation price they must have to supply the product. Paying less than what one was willing to pay or receiving more than one needed results in what economists call an economic rent. It is a return to fixed factors of production, for example, capital and possibly operator's labor.

An ordinary corporation constrained to charge the same single price to everyone cannot capture dif-ferences in economic rent. A cooperative can capture part of these rents by, for example, differentiating membership fees among members in proportion to

u U.S. Department of Agriculture, Economic Research S.rvice, fco11t,,nic lndia.uws of tht Farm &cklr, ECIFS S-2, November 1986.

17

their holdings ofa fixed asset suchas land. By captur-ing some of the rents, the cooperative may be able to cover its costs and still generate net benefits for the membership.

The second key tied to a cooperative's ability to invoke flexible prices is to charge or pay the opli1111ll price. For a purchasing cooperative this price is equal to the incremental (not average) cost of supplying the input or service. For a marketing cooperative, this price is equal to the incremental (not average) net revenue generated. by the raw product (see Figure 5). Revenues generated from charging these optimal prices will usually not cover costs in the type of marketsweare addressing, but they will generate the maximum value for the membership. The coopera-tive then must extract a portion of this value in the form of a membership fee to cover its costs. (Optimal pricing for a cooperative is discussed in Part 2 of this report.)

III. Summary.

The first key to successfully developing an agri-cultural cooperative is that the organization have a genuine economic role to play. The cooperative must be able to provide net benefits for its members in excess of what is available through other market channels, or as the oft-used phase states, "a coopera-tive must be born of necessity."

Part I of this report has been devoted to enumer-ating and analyzing the benefits that a cooperative

may be able to provide in a market-oriented econ-omy as in the United States. Cooperation's possible benefits include the following:

1. Cooperatives may be able to operate more efficiently (on a smaller margin) than nonco-op counterparts.

2. Cooperatives may help farmers avoid the effects of their trading partners market power.

3. By controlling the flow of production or by assuring product quality, marketing coop-eratives may increase prices paid at retail for their finished farm products.

4. Cooperatives may reduce aspects of the risk and uncertainty that plague fanning.

5. Cooperatives may be able to operate successfully in markets that no for-profit company will serve.

Throughout the discussion of these possible benefits to cooperation, particular attention was paid to describing the market structure, conduct or per-formance characteristics, which tend to suggest the typesof market failure thata cooperati vecan correct.

Part 1 of our ''blueprint" is now complete, but even cooperatives that have a critical economic role to play will not succeed unless they are organized, financed, and operated properly. Part II of the report analyzes these keys to successful cooperative development.

18

PART 2. THE ORGANIZATIONAL, FINANCIAL AND OPERATIONAL KEYS TO DEVELOPING A SUCCESSFUL COOPERATIVE

I. Organizational Keys to Success.

This section discusses critical clements in the success or failure of a cooperative relating to mem-bership, vertical integration within a cooperative, and the structure of decision making.

A. Membership.

Probably the two most important causes of fail-ure among cooperatives are insufficientmembership and, hence, insufficient business volume and insuffi-cient equity financing. We discuss membership here and equity generation in the next section.

1. Developing the initial coalition.

Establishing a cooperative first involves devel-oping a coalitionof potential members. But coalitions do not form costlessly. People must invest time and money to bring other proplc together. Herc, how-ever, we encounter a problem with the cooperative organizational form. It offers no special monetary rewards for the people who develop the idea for a cooperative and do the initial work needed to get the business going.

People who perform these roles in developing ordinary businesses are called entrepreneurs. People who perform the entrepreneurial role in successful nonco-op corporations can expect to profit hand-somely from their work because capital (their invest-ment) is the residual claimant in these corporations.

The same type of entrepreneurial vision and energy among key individuals is also needed to develop a cooperative. However, the financial bene-fit to these people is based on their opportunity to patroniz.e (that is, to sell to or buy from) the coopera-tive and will be no different than for any other member. In short, the cooperative organizational form does not offer special financial rewards for entrepreneurship.

No doubt this factor sometimes results in cooperatives not being formed even when there is a need. Even though several farmers may recogniz.e the need for a cooperative, in some cases none may perceive a sufficient individual gain to incur the entrepreneurial start up costs. Two ways exist to surmount this inertia:

1. Appeal must be made to the same sense of civic responsibility that causes people to serve on school boards, town boards, church councils, and other organizations that offer responsibility but little financial gain.

2. As many people as possible should be in-volved in the initial stages to spread the entrepreneurial costs.

Within any communitycerta.inpeople tend to be more civic minded and to perfonn service roles better than others. These people need to be involved in a new cooperative venture as early as possible because the presence of these leaders will stimulate the in-volvement of others and, thus, lower the organiza-tional burden that any one individual must face.

Our survey evidence reported in Part 3 indeed suggests the likelihood of success in a new coopera-tive increases as more people become involved in the initial organizational stages.

2. Building the membership.

As a general rule of thumb, a new cooperative should strive to get as large a membership as possible at the time of startup.The reason is that a vast amount of evidence compiled from years of research in di-verse industries suggests that usually effidency in­creases with size of an apemtion. At worst, efficiency levels arc neutral in respect to size.

Economists use the terms econumies of size or increasing returns lo size to describe the direct correla-tion between size and efficiency. Two types of size economies are distinguished: physical and pecuniary economies. Both are worth discussing in some detail.

Physical economies refer lo the actual produc-tion process, and their presence means that a larger operation can produce at a lower cost per-unit than a smaller counterpart. 1his phenomenon is repre-sented visually by the decreasing portions of the per-uni t cost curves drawn in Figures 2 and 3 in Part 1.

Reasons for physical economies of siz.e are: 1. Economies of mass production including

a. specialization in input use, b. better utilization of capital, c. access to the most advanced technology.

2. Better use of management and central administration.

Physical economies of size are apt to be particu-larly important for capital-intensiveactivities such as fruit, vegetable, meat, or dairy processing and fertil-izer orchemicals manufacturing. The more members the cooperative attracts, the greater its output level will be and the more able it will be to capture econo-mies of size. As the discussion in Part 1 indicated, economies of siz.e are usually exhausted at some outpuI level and the per unit cost curve flattens out as

19

depicted in Figure 2. Only rarely does evidence indi-cate an increasing unit cost curve like the one drawn in Figure3.

Pecuniary economies occur when a larger firm extracts price discounts on its purchases of supplies and price premiums on its sales. Pecuniary economies are important because it is usually cheaper to con-duct business in large, standard-size volumes.

Buyers prefer to make purchases in standard quantities such as the volume of a railroad car or semi-trailer. Sellers who cannot provide these vol-umes will be paid less, if they can make the sale at all.''

Another advantage in cooperative marketing of a large membership and sales volume is that it en-hances the reliability of the product flow. In other words, fluctuations in farm production will tend to cancel across large numbers of producers to result in a steadier flow of product to the cooperative. Experts in marketing invariably cite reliability of supply as a critical factor in establishing a marketing network. Buyers will prefer reliable suppliers and will pay them price premiums.

The key element about economies of size and achieving sufficient membership to exploit them is that the membership must be present at the cooperative's inception. This is due to the dichotomy between what economists call the short run and the long run. When the cooperative is in its pre-operations, planning stage, this is the long run. Essentially the cooperative can at this time adopt whatever size of production and marketing facilities ii feels will most efficiently handle its prospective business volume.

Once this decision is made and facilities are built or acquired, the cooperative is "stuck" in the sense that many of these plant and facilities costs are sunk, that is, nonrecoverable. The scale of operation cannot be adjusted quickly.

Most modem production facilities are designed to operate at a particular rate to efficiently produce a given volume of product. Deviations from this opti-mal volume will cause per-unit production costs to rise usually quite rapidly.

Figure 6 illustrates this point for a hypathetical marketing cooperative. The curve labelled long runis similar to those in Figure 2. Based on its expected membership and the members' expected business volume, the cooperative acquires a plant that can efficiently process Q' units of product per season. However, if members produce more than Q", the plant will be pushed beyond itsefficient capacity and costs will rise along the curve labelled short run.

Figure 6. Per-unit Costs in the Short Run and the Long Run.

Short Run

Long Run

o· VOLUME OF OUTPUT

Similarly per-unit costs will rise rapidly if much less than o· is processed since the facility is operated at Jess than the efficient capacity. In other words, once a facility is acquired, the cooperative will be con-strained to move along the short-run curve.

Development of the appropriate scale of opera-tion is absolutely critical for the success of an emerg-ing cooperative, but making thebest decisions can be difficult. A common cause of recent failures and financial stress among cooperatives has been overex-pansion. Not unlike their farmer members, many cooperatives overinvested in capital facilities during the 1970s. When this capacity turned out to be unnec-essary, the cooperatives ended up reducing volume along the high-cost short-run curve in Figure 6. As such, they were often unable to service the large debt load that usually accompanied these invesbnents..

On the other hand a very conservative coopera-tive may no tend up developing a large enough scale of operation to exploit the available economies of size.

Three axioms for membership policy emerge from this discussion as ways to help cooperatives utilize economies of size while yet avoiding the grave problems associated with over expansion.

1. Subject to geogniphic constraints, member-ship should be made as large as possible at the outset. Although there may be excep-tions to this rule, they are few.

DFor example, evidence of peruniaryeconom.ies in the production and marketing of cotton ;s provided by Edward G. Smith, R. D. Knutsoo, andJ. W. Richardson, '1nput and Marketing Economies: Impact on Structural Change in C.Otton Farming on the Texas High Plains," Amerialn fwmJI! of Agricultural Eamomic:s, 68, August 1986, pp. 716-nO.

20

Some situations are much more amenable to generating a large cooperative membership than others. Needless to say, the greater the economic need for the cooperative, the more likely are farmers to wish to become involved, but other factors can also markedly increase or decrease the transactions costs of building a membership. Organizing costs will be lower when farmers

a. are located in a compact geographical area, b. have similar economic and demographic

backgrounds, and I or c. have a frequent, recurring need for the

services of the cooperative. The first two of these factors influence the ease or

difficulty of obtaining concensus among a group of people. The third factor has been cited by Robert Axelrod in a pathbreaking book on generic coopera-tion (i.e., people getting along) as the overriding feature stimulating cooperative behavior in a wide variety of situations." Evidence to support this point is apparent from examining the market-share statis-tics for cooperatives in various industries in the U.S. Cooperatives' share for frequently-used supplies such as fertilizer petroleum, and chemicals is over one-third. On the other hand, their share for building materialsand farm equipment, important inputs that are not bought frequently, is only about 6 and 2 percent, respectively.is

2. "Wait-and-see" behavior should be discouraged.

Some farmers may be reluctant to join an emerg-ing cooperative preferring to walt and see how it does. Because of the need at the outset to exploit economies of size and peg the correct scale of opera-tion, latecomers are much less valuable to a coopera-tive than are those present at the outset. Membership policies should be structured to reward those who commit at the outset.

3. Membership contracts should be written to encourage long-term commitment of members.

Commitment of members is naturally crucial to maintaining efficient use of the cooperative's opera-tions capacity and to maintain a reliable flow of product in the case of a marketing cooperative. Member commitment is discussed in more detail in the next subsection.

3. Long-term membership policy.

Here we address matters of building commit-ment among members and whether a cooperative should have an open or closed membership.

On the surface, building commitment to the cooperative among members should not be a prob-lem. If the cooperati veissatisfyingeconomic needs of the members, their continued patronage should be assured. Matters are not so simple, though. A well-runcooperative needs to takea long-run perspective to its business which includes investing funds in market development, plant maintenance and/or expansion, research and development, etc.

It is possible, even likely in some years, that noncooperative handlers will be able to offer better short-term deals than a cooperative operating with a long-run perspective. In product marketing this tends to occur during tight-supply, hlgh-price years.

If members do not reliably patronize a coopera-tive, the organization's ability to develop market outlets and efficiently utilize its plant capacity is limited. Thus, just as the cooperative needs to take a long-term perspective, so do farmers with respect to their cooperative membership.

Long-term contracts are one way to encourage member commitment. Five year contracts are com-mon for many large U.S. marketing cooperatives. Of course, contracts can be broken, and it will usually not be prudent for a cooperative to instigate litigation against its wayward members. The alternative is to write the membership contracts so that they pre-scribe the specific levels of patronage that are ex-pected and set forth specific penalties for failing to meet these standards. Patronage standards are best stated in terms of specified volumes of purchases or sales with clauses to release the commitment if, for example, the member experiences crop failure.

Common and appropriate penalties for failing to meet a commitment are;

1. Loss of membership for a specified number of years without explicit approval of the Board of Directors.

2. Loss of rights to payment from the cooperative's revolving fund. (This point is discussed further in the next section on fi-nancing.)

We now tum to address policy towards new members. An open membership is when any farmer meeting certain standards has the right to member-ship; a closed membership is when no new members

~Rob<m A~elrod, ~ E""1ulimro{Coopmilimr, Basic Books, New York, 1984. "Cooperatives' share statistics arc from Charles Kraenzle, "More Farmers Turning lo Cooperatives for Production Supplies, Market- ing.N formerOlq>mUiDts, 49, February 1983, pp. 18-19, aad Roger Wissman, NCo-opShare of Farm Marketing. Major Supply Purchases Climb,N F~rmerC.OOpc•I,,,..., 50, April 1984, pp. 18-19.

21

may join. Cooperatives have had a tradition of open membership. In fact, open membership is sometimes listed as a defining characteristic for cooperatives, e.g., it is one of the so-called ''Rochdale principles."

However, cooperatives do nonually have a legal light to restrict membership, and there is no reason not to use membership policy asa tool to increase the cooperative's chances for success."

From a long-tenn perspective, increasing mem-bership will nonually be good for a cooperative because of the resultingeconomies of size. Reasons to limit membership in the short-tenn are threefold:

1. If the cooperative is operating near its effi-cient capacity, new members would tend to increase operations beyond the efficient capacity, thus raising costs.

2. For a marketing cooperative, the volume of product that can be sold in any season is often limited by sales contracts. Thus, the cooperative may have no sales outlet for the additional product supplied by new mem-bers. At best, new sales mayonlybeachieved at discount prices.

3. New members may not be able to meet ade-quate quality standards.

Quality may relate to the physical quality of the product being delivered or it may relate to the poten-tial member's "quality." For example, if some grow-ers could not be counted upon to reliably deliver the product, it would hinder the cooperative's ability to meet marketing commitments and exploit econo-mies of size. As to product quality, it is difficult and costly in many processing operations to grade indi-vidual producers' products for quality. If everything gets lumped together and the cooperative gets paid based upon the average quality, the high-quality producers end up cross-subsidizing the low-quality producers.

Our research suggests that this problem has been a major source of member discontent in many mar-keting cooperatives. Thus, unless grading schemes for individual production can be conveniently insti-tuted, the cooperative is probably wise and justified in excluding \ow-quality producers.

A fourth, long-term reason to restrict member-ship is when a cooperative is trying to control the

overall flow of product to the market. As discussed in Part 1, this strategy will be effective only if no alternative marketing outlets are available for those denied membership in the cooperative.

4. Nonmember business.

Many cooperatives conduct business with non-members. A recent national survey of fanners indi-cated that about 16 percent patronized cooperatives as nornnembers."

Accepting nonmember business gives a coopera-tive flexibility, and in general, is probably a good idea. Most, but not all of the new cooperatives in our survey were willing to accept nonmember business. In nearly all cases, the reason given was related to better utiliz.ation of capital or simply a matter of needing the business.

The advantage of the flexibility afforded by nonmember business is that thecooperative, by regu-lating its intake of nonmember business, can influ-ence the volume of business it transacts without making a long-tenn commitment. Thus,duling low-supply, high-price years a marketing cooperative would be advised to seek nonmember business as a way of maintaining efficient utilization of plant capi-tal and satisty;ng marketing sales commitments. During high-supply, low-price years, the coopera-tive would not accept nornnember business.

The main risk to acceptingnonmemberbusiness is that it may discourage membership. Some farmers will be unwilling to commit to membership if they can secure most of its advantages as nonmembers while avoiding financing and commitment responsi-bilities. Thus, it is important that the cooperative make choosing membership the more preferred al-ternative to nornnemberstatusfor mostfanners. The reason, of course, is that members with a long-tenn commitment are vital to planning theoptimal scale of operation and securing marketing sales commit-ments.

The two main ways for a cooperative to make membership the generally preferred way of doing business are to

1. offer no guarantees from year to year to accept nornnember business, and

2. pay patronage refunds only to members.

"Good sour<:ei of legal Information 011 orga11izing cooperatives are Israel Padel, The Orgo~iz.Uilm •~d Optratian of ~tiws. American Law Institute, Philadelphia, 1970; Morrison Neely, Ltgol p;,,,,.,, '1 Formn Coaperalive•, U.S. Department of Agriculture, Farmer Cooperative Service, Information Report No. 100. 1976; and James P. Baarda, Siou !llWTp()nltio~ SltlluUs for NrmnCoo,.,..1i,,.., U.S. Department of Agriculture, Agrlrultural Cooperative Service, Cooperative Information Report No. 30, 1982. As lo the question of lbnlling membership, Packet (p. 94) says that membership cannot be arbitrarily denied, but dental can occur if the rooperative can demonstrate that c><istiTig members would incur a lo.s were new members allowed. There are ex"'Plions when membership cannot be denied, for example, when the cooperative provides"" essential service such as electricity. upau! Wilkins, Ml<rb:ling mid Form Supply CooperaliutS: Membership ond USl!, 1980, U.S. Department of Agriculture, Agricultural Cooperative Service, ACS Re.card! Report No. 28, May 1983.

22

Patronagerefundsarethepaymentsmadebythe cooperative to allocate monies after all costs of doing business have been covered. Retaining the income from nonmember business not only tends to encour-age membership but it also is a way of building revenues for the cooperative. Most of the coopera-tives in our survey who accepted nonmember busi-ness did, in fact, retain the income from nonmember transactions.

Federal statutes and some state laws place re-strictions on nonmember business, so before any particular strategy towards nonmembers is adopted, the relevant restrictions must be researched. At the Federal level, several statutes specify that the volume of nonmember business may not exceed that con-ducted with members. Section 521 of the lntemal Revenue Code contains this restriction and also specifics that members and nonmembers be treated the same. Thus, a cooperative that wishes to qualify for section 521 tax status must allocate patronage refunds to nonmembers."

State lawsmay not conform to Federal law on the topic of nonmember business. For example, non-member business is prohibited in some states, while other stales require that patronage refunds be given unly lo members."

B. Optimal vertical integration in a cooperative.

Farmers will observe the effects of market failure at the farm gate, that is, in the level and variability of prices theypa}rorreceive. We saw in Part 1 how such market failure might be corrected by a cooperative. But an important organizational question is how far upstream or downstream into the market flow of Figure 1 must a cooperative intewate in order to correct the market failure?

For example, if farm fertilizer prices appear too high, the problem might becaused byonly oneor two retail sellers in a remote rural area. Alternatively, it might be caused by the market power of fertilizer manufacturers with retailers merely passing on the inflated prices. A successful cooperative must integrate to the stage or stages in the production flaw where lhe market failure is occurring. In the fertilizer example, if the high prices were caused by manufacturer market power, a cooperative organized to sell farm supplies including fertilizer at retail would be ineffective. It would have little choice but to pass on the inflated

fertilizer prices it was charged by the manufacturer(s).

As a very rough rule of thumb, market power is most likely at the manufacturing stages both up-stream and downstream from the farm. These stages are capital intensive with relatively few firms and high sunk costs acting as an entry barrier. Market power at the farm gate may also bea problem in rural areas with only a few sellers or buyers.

Having to integrate back into input manufactur-ing to correct for high prices at retail need not be as discouraging a prospect as it might seem. A number of large cooperatives already operate in these mar-kets and have what is called a fe.derate.d structure. That is, other cooperatives called locals are their members. Thus, farmers who face unsatisfactory prices for re-tail farm supplies may need only form a local retail cooperative which can then apply for membership in a regional or federated cooperative to obtain the right to purchase supplies at cost from the regional.

The key point to this subsection is that market failure at any stage in the production chain will probably be transmitted lhrough the chain to ad-versely affect farmers. A cooperative will success.-fully counteract the market failure only if it is organ-ized to address the market failure. In general this will require integrating into the stage or stages where the market failure occurs. As noted, this integration may be accomplished through a federated local-regional organization.

C. Decision making.

We wish to discuss two points here: voting and management.

1. Voting. Major decisions of the cooperative including the

election of members to the Board of Directors will be made by a vote of the members. Cooperatives tradi-tionally operate on the basis of one person, one vote. Nearly all of the over 60 cooperatives responding to our mail survey employed this system of voting. In fact, some states require use of the one person, one vote system.

However, some cooperatives have adopted more flexible voting systems, and, if the law allows deviation from one person, one vote, such modifica-tions ought to be considered to help attract the larg-est-volume farmers as members. Large farmers can becritical to the success of a cooperative because they provide the lion's share of the business volume

"Section 521 grants certain exemptions from inoome taxation for qualifying cooperatives. The •ecti.on 521 exemptions are in addition to the income-pase-through-to-members provision allowed in subchapter T of the Tax Code and discu9sed in Part 1 here. See Neely and Baarda, Scllrader and Goldberg. and Sexton and Sexton suprR note 3 for more details on the income taxation of OJOpetatives. "See Packel, Neely and Baarda ""P"' note 16.

23

needed to exploit the economies of size in production importance of attaining a sufficient business volume and marketing described earlier. Yet cooperatives to justify a manager's salary. have had increasing difficulty attracting large farm- Another concern of some analysts is that man- ers as members."' agement may not be inclined to run the cooperative

Because a member's financial contribution to a in the member's best interests." The reason for con-cooperative is usually roughly proportional to the cern is that there is no good barometer of volume of business transacted with the a.soodation, management's performance in a cooperative. In large fanners tend to have the largest financial stakes ordinary corporations, the value of the company's in a cooperative. Yet if one person, one vote is used, stock performs this role. The stock price goes up as the largest farmer has the same decision making the company makes more money and vice versa. In power as the smallest. this way, management's pcrionnance can be moni-

The concern is that large farmers will think this tored, stock options can be made part of allocation of power to be unjust and take their busi- management's compensation package as an incen-ness elsewhere." The alternative voting system that tive feature, and poorly performing companies are swings the pendulum in the absolute opposite direc- vulnerable to outside takeover and displacement of tion is to base voting power in proportion to patron- present management-a further incentive to good age levels. This alternative, of course, eliminates any management. semblance of voting power for the smallest growers Because cooperatives either have no stock at all and may discourage them from seeking member- or have stock that is not actively traded, the stock ship. Th.is, too, is potentially troublesome because price cannot be used to induce good managerial obtaining the collective business volume of small performance. This deficiency, in tum, places in-farmers may also be significant to the coopcrative's creased importance on the board of directors in a sucress.n cooperative as a device to insure that management

A voting system that tends to protect both largc- pursues members' best interests. and small-volume members is to grant one vote to A final point conceming management is that everyone and then allow additional votes based on managers should recognize that cooperatives are patronage up to some maximum. With this system different from the ordinary corporations they proba-everyone has some voice, but no one's is too loud in bly were trained to run. A natural inclination is to run the sense of having undue influence. Considering a a cooperative the same way an ordinary corporation systemof this type will be especially important when would be run. there is a large sirediversity among the farmers who This strategy is dangerous because running the might be attracted into a new cooperative. cooperative as an independent "profit center" will

2. Management. not be in members' best interests. The reason is that A cooperative, like any other business, must be the coopcrative's "profits" will come from charging

well run to be successful. Thesimplest way to achieve members too much on their farm supply purchases good management is to hire professional manage- and/or paying members too little on farm product ment with specific expertise, if possible, in the rele- sales. Even if members eventually get the "profit'' vant industry. back in the form of patronage refunds, a cost is still

Our statistical analysis of the reasons for success incurred because economic decisions will have been or failure among the new cooperatives in our survey based on incorrect price signals. Thus, cormnon found the presence of a full-time, professional man- measuresofperformance in noncooperatives such as ager to be one of the most significant predictors of return on equity are not appropriate measures of sucress (seePart3). Ofcourse, professional managers performance in cooperatives. do not come cheaply, a fact which highlights the As Section III describes, a cooperative should be

run so as to jointly optimize the performance of the cooperative and the members' farms. To be able to

"See Wilkins S"P"" note 17. "Some authors have also argued that small farmers may vote for unwise investment proposals at the expense nf their larger col- leagues who will end up footing mos! of the bill through their large patronage volume. See Richard Caves and B. C Peterson. G:lopmllitl<S' 51111~;,, f..,.,,, l"d~slri<s: iftglDIW.liim AM Policy f0<1ors, I !arvard Institute of Economic Research, Di.SCU$Sion Paper No. 974, March 1983. "The fastest growing farrn-•ize classes In the U.$. are the •mallest and the largest. Small farms arc increasing due to the popularity of part-time farms and the increased consumption of fresh fruits and vegetables often grown on small, labor-intensive farms. Both the smallest and the largest farmers are less likely to l>e rooperat:ive members than their intennedlate-slie counterparts. "Pot-example, see John M. Staatz, A Theoretkal Ptrspuli"""" ~ ll<havior of farmers' CoopmHmts, Ph.D. Thesis, Michigan State University, 1984, pp. 106-109 especially.

24

operate a cooperative effectively, professional man-agers must initially come to grips with the funda-mentals ofcooperatives as a distinct business organi-zation. Thus, if management has little prior experi-ence working with cooperatives, members should actively encourage management to participate in cooperative training and education programs."

II. Financial Keys to Success.

In this Section we explore equity generation for a new cooperative, on-going equity acquisition and redemption plans, debt financing. and the use of grants in developing a cooperative.

A. Generating initial equity.

A major characteristic that distinguishes coop-eratives from other businesses is that patronage and not capital is the residual claimant. Thus, members receive income roughly in proportion to their busi-ness volume with the cooperative and not based upon their monetary investment in the association.

This facet of a cooperative's operation tends to make it difficult to generate the equity capital needed to launch a cooperative. The reasons are twofold:

1. The pool of potential equity contributors is limited lo the potential membership. Out-side investors will not find cooperative in-vestments attractive.

2. Even among the pool of potential members, individuals may have incentive to invest as little as possible because their benefit from the cooperative may not be tied to their in-vestment level.

The first reason is due simply to the fact that capital invested in a cooperative earns either a lim-ited return or no return at all. Thus, while an entre-preneur launching a corporation can seek investors world wide, an emerging cooperative must be fi-nanced by the members.»

The possible tendency for underinvestment among members is due partly to a free-rider problem; because capital investments earn little or nothing per sc, members may try to limit their own contributions and "free ride" on others' investments if they can do so and still retain patronage privileges.

A5 we will describe shortly, the free-rider prob-lem can be overcome by tying patronage rights in the cooperative to the provision of equity capital. How-ever, a second problem remains due to the limited resaleability of cooperative investments.

Normally investments in a cooperative includ-ing the right to membership cannot be sold except possibly back to the cooperative at par value. Unlike investors in ordinary corporations who can capture the value of the company's future earnings potential merely by selling their stock, investors in coopera-tives benefit from their investment only as long as they maintain patronage with the association.

Thus, the logical incentive for members is to pressure the cooperative to take a short-run view of the business and maximize current benefits to mem-bers even though such a policy may be detrimental from a long-term perspective. Such behavior is known as the lwriwn problem.

The keys to successfully capitalizing a new coop-erative are, thus, to find methods that overcome members' tendency to free ride and obviate the hori-zon problem.

Once cooperatives are operational, the usual system of retaining a portion of each member's pa-tronage refund or adding a fixed charge for capital accumulation per unit of business transacted (a per­unit retain) does a good job of overcoming the free-rider problem because members' capital contribu-tions are aligned with their patronage levels.

The free-rider problem, though, can be serious when a cooperative is attempting to get its initial infusion of equity. Two alternative systems may be used: Members maybe required to purchase stock in the cooperative, or they may be charged for what amounts to a membership certificate. If a cooperative does not issue stock it is known as a nonstockcoopera­tive as opposed to a stock cooperative.

The legal distinction between the two systems is usually quite unimportant, but we believe the non-stock approach best reflects the economic rationale to join a cooperative which is to secure patronagerights. Thus, if members are required to purchase a member-ship certificate, it is clear what they are purchasing-a right to patronage. Alternatively, requiring the

"Training programs for OO<>p"rative<' directors and managers arc often offered through the oooperative exkmsion oomponent of land grant universities. Other training programs are sponsored by U.S. Department of Agriculture'• Agrlrultural Cooperative Service, and the American Institute fm Cooperation, both headquart...-ed in Washington, D.C. •A relatively recent concept in financing cooperatives that addresses the limited equity pool problem is to involve the «>operative with one OT more limited partners. The partners may or may not be members or employees of the C<X>peratives. The partner• purdiase capital facilities and lease them to the cooperative. The cooperative Is provided the opportunity to buy the partners out afler a specified number of years. The coopcrativetl!lually retain• day-to-day oontrol under these arrangements, but the partner-. also usually have guarantecs, in the form, for example, of a professional management rontract, that the association will be run so a• to protect their Investment. Examples where the partnership concept has been tried include several emerging cable lV cooperatives and Pacific Coagt Producers, a California fruit >md vegetable processor.

25

purchase of noninco~arning stock can confuse members' rights as owners versus their rights as patrons.

To prevent free riding, the membershlp charges need to be made roughly proportional to a member's expected patronage in the association. Some coop-eratives. for example, have charged equal member-ship fees to everyone whlch surely discourages par-ticipation by smaller producers.

The way to make these charges roughly propor-tional to expected patronage is to base them on an Msily observable, fixed asset, which is itself roughly proportional to patronage. Acreage is an obvious proxy for expected patronage in a developing mar-keting cooperative. Information on past marketings or purchases, if available, should also represent a good proxy of future patronage.

Even with Uris system, some members may at-tempt to understate their holding of the asset used to allocate the membership fee. The element needed to overcome this attempt at free riding is to place limi-tations on patronage rights based on the amount of the member's payment for stock purchases or mem-bership certificates. For example, the member who claims to harvest only 100 acres and pays a member-ship charge based on that amount should only be permitted to sell to the cooperative the typical pro-duction from 100 acres unless the member augments his equity contribution.

One final suggestion is to recognize the impor-tance of large producers by making the membershlp fee somewhat less than proportional 10 expected patronage. A simple way to accomplish thls objective is to charge a flat cost to everyone regardless of volume and then an additional fee basedonexpected volume. For example,

Payment= a + b(Acres), in whlch case the fee per acre,

Payment/Acres= a/Acres+ b, declines as acreage increases. Values for a and bare chosen to secure the necessary initial equity infusion. The larger isa relative to b, the greater is the burden placed on small members and vice versa."

The simple fact of thematter is that large-volume producers will often be able to get price discounts on

purchases and price premiums on sales from nonco-operative handlers. Some modifications from basing financing charges purely in proportion to patronage are probably necessary to attract these producers.

John Staatz in hls 1984 doctoral thesis suggests a number of ways to mitigate the horizon problem.17

The three most gennaine to an emergingcooperative are to:

1. Develop a plan to rebate accumulated equi-ties to members as quickly as possible (a topic covered in the next subsection);

2. Lessen restrictions on members' ability to transfer their membership rights. If a mem-bership can be sold with the fann, for ex-ample, its value should become capitalized into the farm's selling price. Also transfer of memberships among generations of the same family should be allowed;

3. Allow members to sell per-unit retain and patronage refund certificates to outside in-vestors.

There are apparently no statutory restrictions on the saleability of membership rights or retained equities in a cooperative. Thus, the cooperative is relatively free to impose whatever restrictions it wishes in its bylaws. Specific concerns with allowing trading of membership rights and equity certificates is that it will raise accounting costs for the coopera-tive and possibly compel the cooperative to incur costly fees to register with the Securities and Exchange Commission.•

11. On-going equity acquisition and redemption.

Once a cooperative is able to begin operations, an on-going source of investment capital is provided by retaining a portion of members' patronage refund or by charging a per-unit capital retain. If members consent, the retained patronage refunds and the per-unitretainsare taxable income to themember and not the cooperative. Because there is usually a net benefit

"Teclmical issues with respect to establishing this type of lee structure are discussed in Pinhaa Zusman, "Group Choice in an Agricultural Marketing Cooperative," G:tuidUm fau""'1 of Ewnomics, 15, May 1982, PP- 220.234, and Richard/. Sexton, "The Formation of Cooperatives: A Game-Theoretic Approach with Implications for Cooperative Finance, Dedsion Making. and Stability," Amn"""1 fou""'1 of Agri<i.-U""'' E<>mDmics, 68, May 1986, pp. 214-225. "See Staatz '"""' note 23. "See David W. Cobia, Roger Wissman, William J. Monroe, Francis P. Yager, and Elmer Purdue, "Equity Redemption: Issues and Alternatives for Farmer Cooperatives," ACS Research Report No. 23, U.S. Department of Agriculture, Agricultural Cooperative Service, 1982.

26

to the member/owners from this tax treatment (see Part 1), the bylaws should include a clause consent-ing to this treatmenl.29

The main problem with this system of equity accumulation is that it may cause discontentment among members who pay income tax onmonies they do not immediately receive and who periodically get noticesofallocationofrefunds that they are unable to convert into income.

Thus, even though patronage refund retentions are a convenient and generally fair way to accumu-late equity, some of their provisions maybe unpopu-lar with members, particularly those who are not very familiar with cooperatives. The concern is that this unrest can induce members to leave the coopera-tive, causing the adverse consequences described in Part 2, Section IA of this report.

Thus, some suggestions are in order to minimize theunpopularityamongmembersoftheequityaccu-mulation programs:

1. Have a visible plan in place to revolve re-tained equities back to members.

2. Be certain that members receive a sufficient cash refund to cover their income tax liability for both the cash and noncash portions of the refund.

In our survey of newly formed cooperatives, the presence of a plan to revolve retained equities to the members was positively correlated with success of the cooperative. Although students of cooperatives have been calling for increased development ofthese plans among ooopcratives,»many cooperatives have no plan in place. Among our sample of new coopera-tives, 62 percent claimed to have a plan in place. However, nationwide it has been estimated tha..t only about one-third of U.S. cooperatives have a system-atic plan lo revolve out retained equities and 29 percent have no plan at all."'

The structure of equity redemption plans is explained in detail elsewhere,"' so only a few points crucial to cooperatives' success will be mentioned here.

Most systematic equity redemption plans refund equity after five, seven or ten years."The idea is that

once a cooperative has acquired a sufficient equity base, it can begin refunding old equities and replac-ing them with new ones. The trade off a cooperative faces is that the more rapidly it revolves out retained equities, the greater the rate atwlrich it must acquire them in order to maintain a sufficient equity base. In other words, a plan with relatively few years from retention to redemption must have a higher percent-age rate of retention to generate the same equity base asa plan with a longer period between retention and redemption.

The specific advantages to having an equity redemption plan are mainly in the area of member relations and include the following:

1. Timely redemption diminishes the horizon problem discussed in the prior subsection.

2. A timely, visible redemption plan will cause farmers to include their prospective refund as part of the benefit from cooperative membership.

3. Receipt of refunds can be limited to those who meet their contractual obligations to the cooperative as discussed in Part 2, Section I.

4. A moritorium on repayments can be insti-tuted by the Board of Directors if financial conditions compel such action. Thus, the mere presence of a plan need not commit the cooperative to payments that will jeopardize its financial health.

As to member tax liability for retained patronage refunds, the main problem occurs when the cash portion of the refund is not sufficient to cover the income tax on the entire allocation. Most cooperative members are sole proprietorships whose farm in-come is taxable on their personal income tax return. Under the current tax law the top personal rates are about 30 percent. Thus, if 30 percent or more of the refund is in cash, the tax liability will be covered and member relations will be improved."

C. Debt finance.

Use of debt has varied widely among cooperatives. For example, in 1962 41 percent of all

"At least 20 ~t of the patron~ge refund must be paid in cash to secure this tax treatment, but the entire amount of the per-unit retains may be retained. There may occasionally be advantages to not qualifying patronage rdunds for taxallon to the members and, instead, paying lax on them al the cooperative. See Jeffrey Royer. "Cash Flow Comparisons of Two Methods of Allocating Coopera- tive Patronage Refunds,0 Paper presented at the annual meeting of the American AgrtcullUtal Economics Association, August 19!!7. "'See Cobia r! al. •wpr11 note 24. ~r. F. Brown and David Votldn, "Equity Redemption Practices of AgricuilUtal Cooperatives,° FCS Research Report No. 44, US. Department of Agrtculture, Farmer Cooperative Servi.,.,, 1977. "'See Cobia •I Ill. '"""'note 28. "Plans that revolve equity only al the time of a member's death or retirement or only at the discretion of the Board of Dirn::tors are not systematic. They do little to solve the problems we have been describing. "See il•prlil note 29.

27

cooperatives had no debt whereas by 1976 that number had fallen to 21 percent.,.

On the other hand, because of cooperatives' equity generation problems described in Section !IA, some cooperatives have placed heavy reliance on debt. Neither of these extreme debt postures are likely to be prudent. However, the optimal debt to equity or debt to asset ratio varies considerably among various industries. For example, in the agri-cultural implement and machinery industry, the median debt to equity ratio is 33.5 percent, while in agricultural chemicals, it is considerably higher, 53.6 percent.

Table 2 reporls the distribution of assels among equity capital, debt capital, and other liabilities (ac-counls payable, proceeds payable, and deferred and accrued items) in1976 for U.S. cooperatives classified by type. A fair amount of variation in reliance upon debt is evident for the cooperatives. Marketing coop-Table 2. Debt and Equity for U.S. Agricultural Cooperatives: 1976

Percentage of Total Assets Represented by

Equity Borrowed Other Type of Cooperative capital capital liabilities

Farm supply 50.6 28.3 21.1 Marketing 35.0 35.5 29.5

Cotton & products 38.2 43.0 18.8 Dairy products 35.4 22.6 42.0 Fruits & vegetables 30.7 44.2 25.1 Grain, soybeans, and products Livestock & wood Poultry products Rice Sugar products Other products

Total

46.9 28.4 24.7 49.7 19.6 30.7 49.2 31.7 19.1 38.4 38.4 23.2 33.9 41.9 24.2 44.5 19.6 35.9 41.7 33,1 25.2

Source: Griffen et al., supra note 35.

eratives tend to rely more heavily upon borrowed capital than do supply cooperatives.

The key criterion in borrowing is, of course, the debtor's ability to make payments on interest and prtncipal. Inability to meet payments is an immediate harbinger of bankruptcy and failure. Reliability of a business's income stream net of variable costs is, thus, the key factor in detennining its ability to take on debt. For example, banks and utilities usually

have among the highest debt to equity ratios because their income streams tend to be very stable. As we have noted, agricultural industries tend to be com-paratively risky, often expertencing wide fluctua-tions in income.

This consideration suggests two policies rele-vant to debt finance:

1. Agricultural cooperatives should rely less heavily upon debt than firms in more stable industries.

2. Whatever strategies can be undertaken to stabilize a cooperative's income stream net of variable costs will both enable it to take on proportionally more debt and reduce the risk of default for any level of debt.

Some of our previous discussion provides the keys to promoting income stability. The main ele-ment is to have a stable membership with committed patronage. Involving the cooperative in marketing or selling several products will also stabilize a cooperative's income stream if the income streams are negatively correlated as discussed in Part 1, Sec-tion IID.

Cooperatives have access to most of the same sources of debt financing as ordinary corporations plus one additional source. The Banks for Coopera-tives (BCsl are lending institutions organized specifi-cally to provide loans to cooperatives.

The BCs are themselves cooperatives; being owned by the cooperatives they serve through capi-tal stock investments. There arc 12distrtct banks plus one central bank. The BCs obtain funds to loan from the sale ofsecuri tics to prtva te investors. Theyare not government funded, but aresupervised as part of the Farm Credit System. This link tends to make their securities relatively safe investments and usually enables the BCs to obtain favorable interest rates which they can then pass on to their borrowers.

The interest advantage that the BCs can offer theirmemberborrowers varies witheconomic condi-tions, but the evidence suggests that a 10 to 25 percent savings relative to commercial bank rates is usually achieved." Thus, the first suggestion relative to debt finance for organizers of new cooperatives is to ex-plore funding opportunities with the BC in their district. Better interest rates is one obvious reason, but a second reason is that the people at the BCs understand cooperativesand may provide an impor-tant source of expertise to an organizing cooperative.

Cooperatives have apparently realized that the BCsare a good lender for them. As Table 3 indicates,

"Nelda Grlffen.Jefhey S. Royer, Roger A. Wissman, Dennis P. Smith, Donald R. Davidson, Steph"11 D. Lury;>, J. Warren Mather and Phillip F. Brown, 'The Changing Financial Structure of Farmer Cooperatives,u Fanner Cooperative Researcll Report No. 17, U.S. Department of Agriculture, Economics S!ll.tistics, and Cooperatives Service, March 1980. "See Caves and Petcnon •WP"' note Zl, appendix C

28

Table 3. Sources of Borrowed Capital for Agricultural Cooperatives: 1976

Percentage of Borrowed Capital Marketing Purcltasing

Source Coop'• Coop's Total'

Banks for cooperatives 72.7 56.1 62.2 Commercial banks 12.4 11.1 9.5 Debt sa-urities 11.6 9.5 18.9 Leases and industrial

revenue bonds 0.9 10.4 3.6 Other cooperatives 0.3 5.6 1.9 Other sources 2.1 7.3 3.9 'Total does not necessarily o::>nform to values for marketing and purdlasing rooperat:ives because of cooperatives that perform both marketing and farm supply functions. Source: Griffen et al. supra note 35.

in 1976 theBCs provided over 60 percent ofborrowed capital for agricultural cooperatives, while commercial banks provided only about 10 percent. Among thenewcooperativesinour study 31 percent indicated use of BC funding while 25 percent obtained funds from commercial banks. The incldence of success, interestingly, tended to be higher among BC borrowers than commercial bank borrowers. The effect was not so strong, however, that we could be assured of its significance in a statistical sense (see Part 3).

Among the other debt sources listed in Table 3, debt securities arc debt instruments issued by the cooperative itself. Normally, only established coop-eratives with a good financial record will be able to use this source of funds.

Industrial revenue bonds do, however, offer another possible source of low cost funds. Industrial Revenue (IR) bonds (also called industrial develop-ment bonds) arc long-tenn debt issued by a local governmental unit (e.g., city or country) to finance the purchase, construction, improvement, or expan-sion of property, plant, or equipment to be leased or sold to private businesses. Alternatively, the funds can be loaned directly to businesses for the above purposes.

IR bonds are not normally tax exempt like mu-nicipal bonds, but they are backed by the stability of the issuing government which enables them to be issued at favorable interest rates with lengthy ma-turities.

Clearly the purpose of IR bonds is to benefit the issuing city or county. Cooperatives, being organiza-tions normally intended to benefit a large number of

community members, are ideal candidates for IR bond funding. The main advantages to the coopera-tive arc the favorable interest rates and lengthy ma-turity. A disadvantage commonly cited is more ex-tensive "red tape" than with other debt sources."'

D. Use of grants and technical assistance.

One perspectively important source of start up funds for new cooperatives is grants. Among our sample cooperatives, 23 percent indicated receipt of at least one grant, and our statistical analysis of the probability of success in new cooperatives does indi-cate a positive correlation between the receipt of grant(s) and success.

Newcooperatives aregood candidates for grants because as small businesses they arc eligible for most funding programs (not, however, Small Business Administration loans) designed to aid the start up of new businesses. In addition, they arc eligible for programs targeted specifically to cooperatives. Funding sources include both Federal and state governments as well as private organizations suchas church foundations. However, eligibility require-ments for programs and the range of available pro-grams thems.elves often change frequently.

The reason grants can be so important to the start up of a new cooperative is that they can effectively provide an additional infusion of equity capital which, as we have noted, is where many new coop-eratives tend to come up short. Grants, thus, provide financial lroerage that can be used to support addi-tional borrowing and, in general, upgrade the organization's infrastructure.

The best advice concerning grants, therefore, is simply this: the range of funding opportunities should be thoroughly explored at the time of initial planning. However, the long-range business plan should demonstrate that the co-op can be successful without continuous grant funding.

Any device, such as grants, that will raise funds or, alternatively, lowercostsduring the planningand start-up phas.e of a cooperative must enhance its chances for success. One way to reducecosts is to take advantage of the network of public sector expertise that can be called upon to advise and assist emerging cooperatives. Sources include the USDA Agricul-tural Cooperative Service, land-grant university ex-tension personnel, farm advisors/county extension agents, and the Banks for Cooperatives. In addition, most states have an agricultural cooperative council-a statewide affiliation of cooperatives-that may provide useful assistance.

"See Donald R Davidson, "Industrial Development Bond Financing for Farmer Cooperalives,q Farmer Cooperative Rescarclt Report No. 18, U.S. Department of Agriculture, Emnomks., Statistics., and Cooperatives Serv1re, 1980.

29

These sources may not always be able to substi-tute for costly private legal, financial, and accounting consultation, but vety often the people involved will have had extensive experience working with coop-eratives and be able to provide extensive advice and assistance at little or no cost.

Among our sample cooperatives, 49 percent reported extensive use of private consultants while the comparable number for the public-sector sources was 78 percent. Our statistical analysis found a posi-tive correlation between use of public-sector consult-ing and success and a negative correlation between private sector consulting and success.

III. Operating Keys to Success.

In this section we di&CUss optimal pricing and output policies in cooperatives and how best to judge the performance of a cooperative.

A. Pricing policy.

If business wisdom could be condensed into one phrase, it would probably be "buy low, sell high." The question is how appropriate is this phrase as a maxim for operating a cooperative? In short, the phrase is half right. A marketing cooperative's goal should be to sell members' production at the highest possible price, and it should attempt to perform whatever marketing functions are profitable in the senseof generating incremental revenues in excess of incremental costs. In essence, on the selling side the marketing cooperative should approximately emu-late the behavior of other marketing sector firms.

Similarly, a purchasing cooperative in its buying activities .will roughly approximate the behavior of its noncooperalive counterparts. It should attempt to procure the supplies it sells lo farmers as cheaply as possible.

What then is the appropriate pricing policy for a cooperative in dealings with members? One sugges-tion contained in the original "Rochdale Principles" is that a cooperative should charge "regular retail prices." In other words, the cooperative would emulate noncooperatives' pricing behavior towards members, too. Cooperatives by their vety nature must breakeven, so an alternative pricing suggestion has been to charge or pay the price that will roughly match revenues with costs. This is known as average cost pricing.

For a cooperative_marketing a single product, Q the break even price,P'"is given by the net-average-

revenue-product curve identified in Figure 5 of Part 1:

PM=( R-C)/V

where R is revenues from sales, C is all marketing costs except raw product costs, and Vis volume of raw product processed.

For a purchasing cooperative selling an input such as fertilizer, we can set price equal to the average or per-unit costs of providing the input or service;

p Total ca;tsforprodudngandselling the product F Amountofproductsold

Average cost pricing will limit the need for pa-tronage refunds, whereas charging regular retail prices will normally result in a profit or surplus that will need to be rebated to members.

Neither of these rules is optimal in the sense of generating the maximum benefit for members from participation in a cooperative. The optimal pricing policy toward members has, however, been known for more than40years." It involves paying members of a marketing cooperative a price equal to the incre­mental or marginal value of their raw product and charging members of a purchasing cooperative a price equal to the incremental or marginal cost of providing the service or supply.

The above rules are known as marginal value or marginal cost pricing. They necessarily maximize members' benefit from participation in a coopcra-ti vc, and, hence, their use can be integral to achieving success in an emerging cooperative. To better under-stand these rules, suppose a marketing cooperative obtains Q, units of raw product from the members. The net revenue, NR, from selling this production in processed form is;

NR(Q,) = Revenues from sales of Q, units - All marketing costs for Q, units except the raw product costs.

Now if the cooperative obtained an additional increment of raw product, it would now process and sell Q, units with the increment, t.Q equal to Q, -Q,. We can compute the net revenue NR(Q2) from selling

units using the above formula. The net value of theQ 2 incremental production is called the net-marginal revenue product (NMRP) and is computed as follows:

NMRP{Q) = NR(Q2)-NRCQ1). •Q

"See Stephen Enke. "Consumer Cooperativ~ and &onom!~ E/f1dency," Amtrico11 Eamomic Rwiew, 35, Marci> 1945, pp. 148-55.

30

NMRP(Q) is graphically illustrated in Figure 5. Our optimum pricing rule says to set price, P .,·,equal to whatever is the value of the NMRP for the given volume of raw product:

pM•=NMRP(Q).

The comparable rule for a purchasing coopera-tive is to compute the incremental costs of procuring an additional amount of the farm input, say Z. Thus, we have AZ = Z, - z,, and C(Z,) = total costs of producing and selling z, units, C(Z,) = total costs of producing and selling z, units. The marginal cost, MC(Z), of the increment is then:

MC(Z) = C(Z2)-C(Z1l •Z

A number of problems may hamper implemen-tation of these marginal cost pricing rules. To begin, while common cost accounting procedures make it easy to compute total revenues and costs or average revenues and costs, the incremental revenues and costs may be harder to discern.

Some simple rules can help overcome this prob-lem, however. A small marketing cooperative can often sell all it wants at a prevailing market price. To get the equivalent selling price for the raw tarm product, we divide the processed product's selling price (PPSP) by the conversion factor, K, as described in equation (1) in Part 1. Finally, we need to compute the incremental processing and selling costs. Fixed costs for plant, administration, and capital equipment by definition do notchange as output varies in the short-run, so they are never part of incremental costs. A rough approximation of the incremental costs is to compute the average varitlble costs, AVC(Q), of proc-essing and selling the raw product:

AVC{Q) = [Production labor costs+ Materials costs +Energy costs+ Other variable costs]/Q.

Our approximate NMRP(Q), and hence, our optimal raw product price, P', is thus:

NMRP(Q) = P"" ,_ (PPSP/K )- AVC(Q).

A similar approximation procedure can be used to arrive at the marginal-cost selling price in a purchasing cooperative.

A second problem is that marginal cost pricing does not directly satisfy cooperatives' breakeven requirement. As Figure 5 illustrates for a marketing cooperative, NMRP is greater (less) than NARP, whenever NARP is rising (falling). Because NARP defines the breakeven price, it follows that pricing according to NMRP will result in a deficit (surplus) when NMRP is greater (less) than NARP.

The analogous results for a purchasing coopera-tive arc that charging a price for inputs or services equal to their marginal cost results in a deficit (sur-plus) if per-unit costs are falling (rising).

This problem is also less vexing than it initially appears. The reason, as noted previously, is that flexibility in pricing is one ofcooperatives' important organizational advantages compared to noncoop-eratives. Thus, what we need to augment the mar-ginal cost prices described above is an additional set of charges or rebates to enable the cooperative to balance its books.

If marginal cost pricing is expected to result in a deficit, the additional necessary revenue can be ob-tained from annual membership charges structured according to the criteria discussed in Section IlB. Surpluses can be refunded through the usual system of patronage refunds, but one problem is created. Rebating strictly according to patronage causes the pricing scheme to degenerate into average cost pric-ing which as we have noted is not optimal. The preferred alternative would be to return surpluses in the form of a rebate on the membership costs where the membership fee is set up as described in Section JIB.

The final problem with implementing thepricing policy described here isdue to people's general igno-rance concerning the distinct organizational and operational features of a cooperative. Potential credi-tors, in particular, may expect a cooperative to oper-ate similarly to ordinary corporations. They may be unwilling to loan money to a cooperative that openly expects to accrue a deficit to be covered with a set of membership charges. If in this sense the world is not ready to accept cooperatives being operated as they should be, a cooperative may have to modify its behavior to fit others' preconceptions.

B. Output policy.

This section is concerned mainly with the ques-tionofwhether a marketing cooperative should try to actively influence its members' production levels, or, instead, passively act as a "home" for whatever the members produce. We begin by noting that im-proved communication and information flow is one of the organizational advantages of cooperatives discussed in Part 1. Thus, at minimum, cooperatives ought to employ this advantage to communicate information about market conditions to members and, in tum, to receive information from members on production plans, harvest schedules, etc.

Coordination between the cooperative and the membership as to desired product characteristics and harvest times can both improve quality and cut

31

costs by promoting efficient use of processing can be sold only at a substantial discount, the sales facilities. limitations should be communicated to members via

Whether a cooperative should pursue a more active role in regulating members' production de-pends on a number of factors:

1. Does the cooperative have any significant influence on the prices it receives?

2. Is the cooperative restricted as to the amount it can sell at a given price?

3. Can the cooperative keep product off the market?

4. Is plant capacity a concern? As we noted in Part 1, marketing cooperatives

mayhave some influence on their selling price if they control a sizable share of production in a local or regional market. When markets are national or inter-national in scope, influencing price is possible only for the largest, most powerful cooperatives.

If a cooperative does market a significant shareof production in the relevant market, its opportunity to influence price depends on its ability to keep product off the market. Merely limiting the amount of prod-uct it will accept from members will not raise market prices if the members can turn around and sell their surplus production through other channels.

For example, cooperative cotton gins often sell cotton seed to milling cooperatives through a feder-ated structure. The mills buy seed from the gins and process it into oil and meal or sell it as whole seed. However, many gins also sell whole seed directly. Obviously, any attempt by the mills to regulate the flow of whole seed would be rendered ineffective in these markets due to the selling activities of the gins.

Thus, our first occasion to consider restricting member deliveries is when the cooperative can raise price by restricting output and can keep the surplus production off the market. Ofcourse, the cooperative must communicate to members in advance the vol-ume of production it will accept from each. Members can then divert resources into other types of produc-tion.

Although it may seem harsh to deny members a home for their production, it is important to under-stand that the price increases obtained can more than compensate them for the lost production. It should also be understood that volume restrictions must be made binding to accomplish their intended goal. Voluntary programs will be rendered ineffective by free riders.

A second occasion to restrict deliveries is if the cooperative's selling opportunities are limited. In some industries, sales are governed by long-term contracts, so a cooperative's selling opportunities may be known well in advance of production. In these cases, if surplus production cannot be sold or

restrictions on their raw product deliveries. These first two reasons to restrict member deliv-

eries are demand related. The final reason is due to production costs. Modem plants, especially capital intensive ones, are geared to efficiently operate at a given volume level. Deviations from the efficient volume cause costs to rise rapidly as illustrated in Figure6.

If the cooperative has reached its efficientoperat-ing capacity, it should restrict member deliveries in the short-mn until it can add additional capacity.

The point with any of these reasons to deny members a "home" for their production is that such a denial is in their mutual best interests. Tf delivery restrictions are clearly communicated in advance, members can either find altemativemarket outletsor alternative products to produce. Meanwhile, their net revenue per unit from the cooperative will be higher than if the surplus production had been dumped at the cooperative's door.

Acting as a "home" for members' production is good member relations and, thus, it should be the prevailing policy, unless one of the three reasons cited above is compelling.

C. Evaluating a cooperative's performance.

Once a cooperative has been developed, the members must judge its performance to determine if the organization is, in fact, accomplishing the objec-tives that led to its formation. The question for members individually and collectively to ask is, "am I (are we) better off with the cooperative than without it?" This question can only be answered from a reference point, and two are available;

1. the members' economic well-being before the cooperative was developed;

2. the economic well-being of fanners who chose not to join the cooperative.

The key point about either of these comparisons is that they ordinarily mustbemade from a long-term perspective. Only in a few isolated instances will the advantages or disadvantages of a cooperative be immediately apparent. Reasons why a Jong-term evaluation is needed are that;

1. Most businesses, cooperatives included, do not perfonn at peak potential in their first year or two.

2. Many factors completely unrelated to the cooperative can cause farm prices to rise or fall. Thus, to avoid erroneously attributing random price effects to a cooperative, mem-bers must evaluate price performance over

32

the long-run when random shocks should average out.

3. In the first few years of operation the cooperative will need to retain a large portionofearnings for investmentpurposes. Moreover, any plan to rebate retained equities will not have ''kicked in" yet. Thus a cooperatives' price performance may look misleadingly poor in the initial years.

Another factor to bear in mind when judging a cooperative' s perfonnance relative to a noncoopera-tive counterpart is that the presence of the coopera-tive may have caused other firms in the industry to behave more competitively, that is, to offer higher prices and/or charge lower prices. This possible procompetilive effect ofcooperativeshasbeen called the competitive yardstick, and it has an element of free ridership attached to it; nonmembers effectively may gain many of the benefits of the cooperative due to the noncooperative handlers' more competitive behavior.

Jn spite of this free rider problem, it is clear that members probably could not disband the coopera-tive and revert to patronizing the other handlers because the competitive yardstick would no longer be in place.

Notice that in this discussion of cooperative perfonnance evaluation, we have stressed the mem-bers' long-tenn economic well-being compared ei-ther to nonmembers or the pre-cooperative era. We have not mentioned anyof the indicators such as rate of return-on-equity (ROE) commonly used to judge businesses. These indicators are invalid to judge a cooperative's performance even though they have frequently and erroneously been used for that pur-po~.

The reason is that it should not be the cooperative's goal to accumulate vast amounts of "profit" and thus, to achieve a big ROI. Profit, of course, would normally be achieved at members' expense by charging too much for supplies and/or paying too little on members' raw product sales. Such behavior asoursubsection on pricing has indicated is not in the members' best interests.

Rather, as with any vertically integrated organi-zation, the revenues and costs of the cooperative are accountable to the members, and the well-function-ing cooperative operates specifically to benefit its members. Thus, the cooperative's perfonnancemust be judged based on the members' welfare after they have received its benefits and paid its costs.

IV. Summary.

Part 2 of our report has focused on the organiza-tional, financial, and operational keys in successfully developing a cooperative once a reason for its exis-tence has been established (Part 1).

Among the organizational keys, attaining as large a membership as possible was cited as an espe-cially crucial factor in enabling the cooperative to exploit economics of size in both production and marketing. Suggestions were made for specific fi-nancing and decision making frameworks that would enable the cooperative to appeal to both large and small farmers.

In thearea of financing, we stressed that contrib-uting equity to a cooperative was often not an attrac-tive investment per se. We suggested characterizing initial equity contributions as membership fees and further suggested that patronage rights in the coop-erative be tied directly to these fees.

As to operations, we stressed that the key feature was to set up price and output policies so as to jaintly optimize the performance of the cooperative and the member fanns. Optimal pricing policies were shown to be those that paid members the marginal value for their production or charged them the marginal c05t of providing supplies. Also, several instances were isolated wherein a policy of restricting member de-liveries to a marketing cooperative could be benefi-cial.

This completes our "blueprint" to success in developing a modem agricultural cooperative. To test and apply the blueprint we tum now to examin-ing the experiences of a number of recently devel-oped American agricultural cooperatives.

33

PART 3. AN EXPLORATORY ANALYSIS OF RECENTLY FORMED AMERICAN AGRICULTURAL COOPERATIVES

This part of the report describes our survey analysis of the experiences of many recently-formed American agricultural cooperatives. Section I de-scribes our methodology to locate new cooperatives and develop contacts; Section II diS(;usses the survey instrument; Section III reports the main survey re-sults, while Section IV contains a statistical analysis of the determinants of success in new cooperatives.

I, Identifying Newly Formed American Agricultural Cooperatives.

To provide ideas and a fertile testing ground for our cooperative success blueprint, we sought to lo-cate and develop contacts with cooperatives formed in the U.S. and affiliated areas since roughly 1970 onward. This task was complicated because no cen-tral directory of these organizations is available. Moreover, many new cooperatives are quite small and not well known.

Our procedure in developing a sample of re-cently formed cooperatives was to contact and re-quest assistance from individuals and organizations most likely to be aware of recently-formed agricul-tural cooperatives in the various parts of U.S. Those contacts included:

1. Personnel at land-grant universities with primary responsibilities in the area of cooperatives;

2. Directors of state-level agricultural councils; •nd

3. Heads of the 12 district Banks for Cooperatives.

Each of these individuals was sent a letter in-fonning him/her of our study and requesting the names, addresses, and contact persons for coopera-tives formed since roughly 1970. Because of our interest in success and failure, we were particularly interested in finding organizations that had begun operations since 1970 but had subsequently gone out of business.

The response rate to this request was high, but several people were unaware of any newly-formed cooperatives in their area. Nonetheless, through this network and other contacts, we were able to locate about 150 cooperatives in the U.S. and affiliated territories that met the criteria for inclusion in our study.

Our procedure was then to make telephone contact wlth one or more leaders of these coopera-tives, explain the purpose of the study, and request permission to mail a survey instrument to be com-pleted by the person. The leaders we contacted were either members who were on or had been on the Board of Directors (usually in the capacity of presi-dent) or were professional management. Among our respondents, 64 percent were member/patrons of the cooperative and 36 percent were from manage-ment.

Contacts were generally willing to participate, and surveys were mailed topcopleaffiliated with 108 different cooperatives. If the survey form was not returned within four weeks of mailing. a follow-up letter and an additional survey form was mailed. 1bis methodology produced a response rate of nearly 60percent.

II. The SW"Vey Instrument.

Because our survey encompassed both purchas-ing and marketing cooperatives and both active and out-of-business cooperatives, four survey instru-ments were designed, one each for active marketing cooperatives, active purchasing cooperatives, no-\onger-operating marketing cooperatives, and no-longer-operating purchasing cooperatives.

The methodology involved designing a prelimi-nary set of survey instruments and submitting them to extensive pretesting from case study analyses of several recently-formed California agricultural co­operatives. Personal interviews by either or both of the authors were used in the pretesting phase. Based on the availability of cases in California, the pretest-ing was skewed towards cooperatives marketing fresh and processed fruits and vegetables. Both active and defunct cooperatives were included in the pretesting.

Final versions of the questionnaires were pre-pared based upon results from the pretest phase. Cooperatives used in the pretesting phase were not included in the final survey phase for which results are reported here.

Category areas for the surveys were chosen to test, illustrate, and, as appropriate, modify or update the cooperative success blueprint reported in final form in Parts land 2of this report. Survey questions included the following categories:

34

1. Basic information on the cooperative including age, membership, products marketed, and position of the respondent;

2. Economic factors motivating the decision to develop the coopcra tive;

3. Membership factors; 4. Financing issues; 5. Decisionmaking and management; and 6. Evaluation of the cooperative's success or

failure. A copy of the survey instruments mailed to ac-

tive marketing and active purchasing cooperatives is included at the end of the report. Survey forms sent to leaders of no-longer-active cooperatives differed only slightly to reflect the ex post nature of these organizations. Where judgmental answers were re-quested, we allowed respondents to choose from among three categories: For example, A. an impor-tant factor, B. a minor factor, C. not a factor. When-ever appropriate, we encouraged respondents to elaborate upon their answers and left space on the form for that purpose.

m. The Main Results.

A. Characteristics of survey respondents.

Sixty one usable responses were received. Re-sponses emanated from 26 states plus Guam and the Mariana Islands. Hawaii was the most frequently represented stale with eight responses, followed by Vermont with six, and California, Minnesota, and Tennessee with five each. A complete distribution of responses by state is provided in Table 4.

The most frequent start updates for our respon-dents were 1983 and 1984 with eight beginning op-erations in each year. The complete distribution of start-up dates is provided in Table 5.

As to the types of cooperatives responding, the most frequently represented function was fruit and vegetable marketing with 22 responses (36 percent} with farm supply cooperatives second most repre-sented with 11 responses. The complete distribution of the sample cooperatives by function is reported in Table6.

Locating leaders of cooperatives that had gone out of business who were willing to participate in the study was, not surprisingly, difficult. Eight re-sponses from tills group were obtained with the remaining 53 coming from cooperatives that were active at the time of the study.

Survey response data for the four main compo-nent areas of the questionnaire: economic factors, membership factors, financing, and decisionmaking, are now presented and analyzed in tum.

Table 4. Cooperative Survey Respondents by State or Other Location

Number of State/Location Responses

Alabama 3 Alaska I Arkansas 2 California 5 Florida I Gu= I Hawaii 8 Idaho I Kentucky 2 Louisiana 2 Maine I Mariana Islands I Minnesota 5 Missouri I Nebraska 2 New Mexico I North Carolina I North Dakota 2 Oklahoma I Oregon 2 Texas I Tennessee 5 Vermont 6 Virginia 3 Washlngton 2 Wisconsin I Tutal 61

B. Economic factors. Part 1 in this report documented several eco-

nomic benefits that cooperatives can prospectively provide their membership. Question 7 on either survey form was designed to capture the importance ofeachofthesefactors: Question 7a, "prices were too low (marketing ro-op) or too hlgh (purchasing co-op)," was designed to mainly capture market power considerations (Part 1, Section IIB). Question 7b on the marketingcoopcrative form, "marketing services were not performed effectively," reflects the possibil-ity that the marketing margin may be reduced through cooperative marketing (Part 1, Section !IA}. Question 7b on the purchasing cooperatives' form reflects the same idea on the farm input side of the market.

35

Table 5. Start-up Dates for Cooperative Survey Respondents

Number of Date Responses

Pre 1970 4 1970 1 1971 1 1972 4 1973 2 1974 4 1975 4 1976 4 1977 2 1978 3 1979 2 1980 3 1981 3 1982 4 1983 3 1984 8 1985 8 1986 1

Question 7c, "prices were subject to too much variability or uncertainty," captures the possible risk-reducing roleofcooperativesdiscussed in Part 1, Section IID. Question 7d on both forms reflects cooperatives' possible role when no private handlers will operate in a market (Part 1, Section IlE) or the similar problem discussed in Section IID that buyers and sellers may be present but cannot be counted upon to meet farmers' buying and selling needs. Question 7e on the marketing form, "individual farmers lacked bargaining power," is intended to capture both farmers' difficulties in dealing with monopsonyoroligopsony buyers{Part 1,Section IlB) and their aspirations to raise market prices through

Table 6. Cooperative Survey Respondents by Function

Number of Cooperative Function Responses

Fruit and Vegetable marketing 22 Farm supplies 11 Animal products marketing 8 Farm services 5 Nut marketing 4 Grain marketing 4 Other marketing 7 Total 61

collective marketing (Part 1, Section IIC). Space was also provided for respondents to list and rank addi-tional factors.

Respondents were asked to indicate whether each of the above considerations was (A.) an impor-tant factor, (B.) a minor factor, or (C.) not a factor in the decision to organize a cooperative. The quantita-tive responses to these questions are reported in Table 7. In addition to indicating the percentage responses to the A, B, and Ccategories, the table also reports a mean or average rating based on assigning the values 1.0, 2.0, and 3.0 to A, B, and C responses, respectively. Thus, factors with a lower mean score were judged to be relatively more important than factors with a higher mean score.

The first point to note from the table is that all factors weredeemed tobe very important orofminor importance by over 60 percent of the respondents. The most important factor for marketing coopera-tives wasbargaining power which 70 percent cited as a major factor underlying !heir organizing effort. Representative respondent comments about the bargaining power factor included:

Table 7. Importance of Alternative Economic Factors in the Decision to Form a Cooperative

Important Minor Not a Me~

Factor factor factor factor response

-percentage responses--7a: Prices too low (high) 44.3 23.0 32.8 1.89 7b' Ineffective or poor quality

supplies or service 54.1 16.4 29.5 1.75 7c: Price variability I

uncertainty 37.7 26.2 36.1 1.98 7d: Undependable/nonexistent

market outlet 65.6 9.8 24.6 1.59 7e: Bargaining power

(mtg. co-op only) 70.0 10.0 20.0 1.50

36

•Produce buyers found it easy to pi tone fann against another; farmers had no clout. •Individual farmers lacked quantity to bargain effectively. •Buyers would tend to take advantage of individual farmers.

Among the fann supply and services coopera-tives, the most important factor was undependable or nonexistent source of supplies which 65 percent rated as a very important factor in the decision to organize a cooperative. Representative comments included:

•Plantation stopped selling fertilizer and herbicides to growers. •No other almond hullers in this area would handle the new producing trees. •No supplies/services on this island. •Tractor services were not available.

Although respondents commented extensively in the open response categories ( 7£ and 7g on the marketing form, 7e and 7f on the purchasing form), their responses tended to be restatements of the same basic economic problems suggested in the main re-sponse categories.

C. Membership factors.

We were concerned with membership growth from initial planning to start up to current size, the decision to open or close the membership, and policy towards nonmembers.

The response to question 4 concerning the num-ber of producers involved in the initial planning stages confirmed the hypothesls that a few people often end up doing much of the costly preliminary work in a cooperative. Ten or fewer people were involved in 54 percent of the cases and 20 or fewer were involved in 79 percent.

To measure the growth of the membership from the time of initial planning to start up, a growth coefficient (GC) was computed as the ratio

GC Members involved in initial planning stage. Members at the time of initial operations

The mean value for the growth coefficient in our sample was 0.65, indicating that onaverage the coop-eratives in our sample failed to double in size as they evolved from planning into an actual operating en-tity. Our statistical results reported in the next sec-tion demonstrate that the smaller growth coefficient (i.e., the greater the growth rate) the greater the prospects for success in a cooperative.

Among our sample cooperatives, 72 percent reported having an openmembership (question 16a),

Table 8. Reasons for Open or Closed Membership in a Cooperative

Important Factor

Minor factor

Not a factor

Mean factor response

Open membershlp: --percentage responses---

More members make co-op run more efficiently 60.7 14.8 9.8 1.40

More members give co-op more bargaining power 60.7 9.8 "' 1.46

Open membership is the co-op way 54.1 13.1 18.0 158

Closed membership:

More production would cause operating inefficiencies 625 0 375 1.75

Restrict volume of product on the market 16.7 0 83.3 2.67

37

while 28 percent indicated otherwise. Open mem-bership is oneofthe Rochadale principles, but, as was shown in Section IA of Part 2, open membership may not always be the most prudent membership policy. Thus, we were interested in the cooperatives' reasons for having an open or closed membership policy.

Question 17 on both forms was designed to generate this information. For open membership cooperatives, the following factors were offered: more members make the co-op run more efficiently (a physical economies of size factor), more members give the co-op more bargaining power (the market power argument)and open membership is the co-op way of doing business (the Rochdale principle).

Responses were generated using the (A.) impor-tant factor, (B.) minor factor, and {C.) not a factor categorization. Percentage responses are summa-rized in Table Balongwith the mean response based, once again, on the A= 1, B = 2, C = 3 scale. Over 60 percent of the respondents thought that both the economic reasons to keep membership open were important factors, but, interestingly, the Rochdale principleofopen membership as a matter of coopera-tive philosophy was found to be important by 54 percent of the respondents.

Representative comments of respondents con-cerning the rationale for open membership included:

•We have a limited market; more members gives us more price control; more volume gives us credibility. •We opened membership when we realized that new members did not have an adverse effect on existing members' (sales]. •More members means more .. .income for advertising and promotion. •Co-op can most effectively control and stabilize priceifmost growers arc members. •A large co-op has a better buying power than a smaller one.

Potential reasons to close membership listed in Part 2, Section IA included plant capacity limitations, quality control, and desire to restrict the flow of producton to the market. Because most of the sample cooperatives had open membership, response to the closed membership question was limited but, none-theless, illuminating. Over 62 percent cited plant capacity limitations as an important factor in their decision to dose the membership, while only about 17 percent cited a desire to limit access to the market. This result, in tum, affirmsour earliercondusion that volume restriction is unlikely to be feasible for an emerging cooperative.

Representative comments about the decision to close membership included:

•Havingjust started operations, the demand for service exceeded our capabilities. •Processing plant currently at full capacity; capital costs for expansion too costly for anticipated return from product. •Transportation services offered now to all in the area; more expansion lo off line shippers would not help our situation.

Turning now to theissueofnonmember business (questions 18 and 19), 72 percent of our sample coop-eratives accepted nonmember business and 28 per-cent did not. Among those accepting nonmember business, 75 percent retained the income and paid taxes upon it, thus, effectively using the nonmember business as a source of profits. (The decision about whether to accept nonmember business is discussed in Part 2, Section IA4.)

Respondents commented extensively on the reasons for accepting nonmember business with the responses invariably pointing to either of two sound business practices:

1. Using nonmember business to increase plant efficiency, i.e., to exploit economies of size;

2. Using nonmember business to meet sales commitments or unanticipated sales opportunities.

Representative comments on the efficiency argu-ment included:

• Nonmember business brings us to a level of efficient facility utilization. • More volume would lower unit costs.

On the idea of using nonmember business to generate flexibility in marketing cooperatives' sales, the following were representative comments:

•Because sometimes we can use [nonmember business] to fill a gap and maintain a market. •To meet contract commitments that members cannot meet. •Co-op will accept produce only if not available from members. •If we need it to satisfy C'Ustomers' needs. •If we can make a profit on nonmember business, we do it.

0, Financing.

We were interested in several aspects of financ-ing: sources of initial equity and debt capital, extent of use of grants, and whether a plan had been estab-lished to revolve retained equities back to members.

Thirty nine of the responding cooperatives (64 percent) indicated use of membership fees or assess-

38

ments to generate initial equity capital, while 26 (43 percent) reported issuing stock. Purchasing or sup--ply cooperatives more frequently chose the stock approach than did their marketing counterparts. Several cooperatives used a combina lion of member-ship fees and stock issuances. We also attempted to generate information on the new cooperatives' sourcesofadditional equity capital beyond the initial infusion. However, response to our question 20 was poor, probably because an accurate answer would have required access to financial statements, and, therefore, no worthwhile inferences could be drawn from it.

As to initial debt capital, 25 percent reported borrowing from commercial banks, and only 19 {31 percent) reported borrowing from the Banks for Cooperatives {BC); even though they usually offer somewhat lower interest rates than commercial banks. One possible explanation for the rela ti ve!y low usage of the BCs is that they may sometimes be unwilling to accept risks of loaning to new coopera-tives. Interesting to note is that 15respondents Hstcd access to debt financing, especially the BCs, as a veiy important factor in choosing a cooperative organiza-tion rather than an alternative organization form (questions 8 and 9). Thus, even though the propor-tion borrowing from a BC was not high, most of those who did obtain BC funding attached considerable importance to it.

Nineteen respondents (31 percent) reported ob-taining loans from alternative sources. Among those who specified the source, four reported borrowing from members, and five reported receiving loans from state agencies. Five {8 percent) of the sample cooperatives reported having no debt at all.

Only14 cooperatives(23 percent) reported u.se of grants as an initial source of funds. Our survey was unable to discern whether failure to obtain grants was due to failure to apply or to applications being rejected. The grants received came from a wide variety of sources:""

Federal agencies 5 State agencies 4 Tennessee Valley Authority 4 Private donors 2

The Tennessee Valley Authority's support was, of course, limited to cooperatives in its Southeastern U.S. service area. All in all the fairly limited use of grants relative to the number of possible available sources indicates that grantsmanship is an area where leaders ofnew cooperatives may need to focus greater attention.

Thirty eight respondents {62 percent) indicated that the cooperative had a plan in place to revolve retained equity back to members. This figure is probably overstated to some extent because some respondents appear to have confused the payment of patronage refunds with the presence of a revolving fund. The most commonly described plan (eight cooperatives) was nonsystematic and based on the year-to-year discretion of the Board of Directors. Two cooperatives redeemed equities based on a member's age, 65 and n, respectively.

Among the various systematic plans described, the time from acquisition torevolvementof the equi-ties varied fairly evenly from four to tenyears. Some cooperatives with lengthy cycles, e.g., lOyears, indi-cated plans to speed up revolvement when the cooperative's financial health pennitted it. Among the cooperatives with no plan, only a few indicated an intention to develop one.

E. Decision making and managemenl

Our concerns in this area were with the informa-tion and planning that went into the decision to develop a cooperative and whether any alternatives to a cooperative were considered. We were also interested in the nature of voting in the cooperative and its management structure.

As noted in Part 2, Section !ID, a prospective advantage to starting a cooperative is that several public sources ofinformation and expertise areavail-able to lend assistance. These include university extension specialists, the USDA Agricultural Coop-erative Service, farm advisors or country extension agents, and Bank for Cooperatives personnel. Pri-vate consultants,of course, arealso available for a fee.

Our question 10 was designed to indicate the extent to which our sample cooperatives made use of the various information sources. Table 9 summarizes the results based on the A. important source, B. minor source, and C. notasourcecategorization. The mean score reported in the table results from assigning A= 1.0, B = 2.0, C = 3.0.

The most important information source was university extension specialists which over one-half listed as an important source. However, private consultants ranked nearly as high, being cited as an important source by nearly one-half of the respon-dents.

Those reporting the BCs to be an important infor-mation source coincides almost exactly with the number of co-ops reporting BC loans. This result

"Not a!l respondents listed the grant source, and some listed multiple sour""9.

39

Table 9. Information Sources Used in Developing New Cooperatives

Important Minor Not a Mwn Source factor factor factor =re

University extension specialists 52.5 -percentage responses-

9.8 37.7 ,,, Agricultural Cooperative Service 21.3 23.0 55.7 2.34 Farm advisors/county extension agents 41.0 19.7 39.3 1.98 Bank for Cooperatives 29.5 8.2 62.3 2.33 Private consultants 49.2 14.8 36.1 1.87

affirmsour point in Part 2, SectionIIC that,aside from providing low-cost loans, the BCs can also provide a source of low-cost expertise.

Frequent use of expensive private consultants relative to the publicly available services which are usually free suggests that some of the new coopera-tives may have been overlooking the best values in terms of information and expertise.

Finally, to determine the extent of formal plan-ning for the cooperative, we asked in question 11 whether a financial feasibility study had been con-ducted. The answer was nearly evenly split: 31 indicating "yes," 30 reporting "no."

A cooperative is not necessarily the only organ-izational vehicle through which farmers can address the various elements of market failure discussed in Part 1 of this report. For example, farmers who face market power in their sales markets could in prin-ciple acquire and operate a marketing firm as either a partnership or a corporation. Special subchapter S corporations are allowed to remit corporate income back to their shareholders for taxation in a similar manner to the tax treatment afforded cooperatives. Question 8 on our survey was intended to see if any of these organizational alternatives had been consid-ered. Mostly the answer was "no." Only three reported considering a partnership, only three con-sidered the subchapter S mode, while 17 considered an ordinary corporation.

The reasons for choosing the cooperative form were requested in question 9. Although this question was to be answered only by those who actively considered alternative organizations, the number of responses to this question was somewhat greater than for question 8. Access to BC funding was listed

as an important factor by 15 respondents, while 14 listed tax considerations. Note that partnerships and subchapter S corporations are taxed similarly to cooperatives, but ordinary corporations would be subject to double taxation (see Part 1,Section IIA). Six respondents cited the prospe.:tive membership size as an important factor. Whereas ordinary corpora-tions and cooperatives can have any number of shareholders, shareholder numbers in parlnerships and subchapter S corporations are restricted by law."'

Voting in our sample cooperatives was almost exclusively done on a one person, one vote basis.. Only two respondents reported voting in proportion to patronage. This situation is an instance where the prevailing practicediffersfrom recommended policy contained in our success blueprint. Unless all mem-bers are similar in their volume of business, one person, one vote systems threaten to causediscontent among the larger members who arc crucial to a co-op's success. However, recall that one person, one vote systems may be unavoidable consequences of state law in some cases.

As to management (question 23), 21 respondents (34 percent) indicated that the cooperative was managed by the producer/members themselves, 36 (59 percent) indicated the presence of full-time pro-fessional management, while 6 (10percent) reported used part-time professional management. (The per-centages add to slightly more than lOOpercentdue to multiple answers on a few of the fonns.) Our statis-tical analysis of the determinants of success in a new cooperative suggests that the presence of full-time professional management is an important key to success. We tum to that discussion now.

•A subchapter S corporation can have no more than 35 sharchold~rs. Only individuals, not corporations, partnerships, etc. may be •harcholden In a subcl>apter S corporation. There appears to be no strict legal maximum to the number of partnen; In a partnership, but the Jaw requires that the =-owners intend to .u!i»rly J"'rlU:ipllle in the trade or \>usiness. This proviso dearly reflects a presumption that partnerships will involve a small number of c0<01'"Iler$.

40

IV. Statistical Analysis of the Determinants of Success in Emerging Agricultural Cooperatives.

A. The success ratings.

The survey respondents were asked to rate their rcccntly-fonned cooperative as either a major suc-cess, a minor success, too early to tell (about success}, or not successful (questions 28 and 31 on the market-ing co-op and purchasing co-op forms, respectively}.

The results were as follows:

Number Percentage

Major success 28 45.9% Minor success 12 19.7 Too early to tell 10 16.4 Not successful 11 18.0

All of the cooperatives that had ceased operating were classified as not successful.

Representative comments from among those who ranked their cooperative a major success included:

•We have organized 50+ small growers into a singlemarketingforce. Sales are increasing about 25-30 percent per year and membership is growing annually. We have circumvented the market by going direct to chain warehouses. •We have had two solid years with good prospects for coming years. Membership is increasing. For each of our two years of operation, we have returned 20percent more to our growers than the independent field price. • The co-op has fulfilled its purpose for being organized-toprovide goods and services to its members at competitive prices and return its profits back to its members. • In ten years the co-op has repaid all original capital loans. Has good volume, quality members, good equipment and personnel, short revolving cycle, etc.

Among the comments of those less sanguine about their cooperative's success included the following:

•Wearegetting by. Co-ops are complex like the people they comprise. • I believe we will find two major problems: (1) trucking has cost us too much because

production was down due to the weather, and (2) it is too much for a farmer to handle management. • Possiblenewmembersareina wait and see attitude, and the five year commitment they mus! make to sell all their cattle through the co-op is hard to do.

B. A statistical model of the determinants of success.

1. Constrncting a measure of success.

No business's success is guaranteed. Bad luck, unexpected adverse economic conditions, new com-petition, changing consumer preferences, etc. can topple even the most carefully planned, best run enterprise. However, the main purpose of this report has been to suggest that thercareparticular economic conditions and certain organizational, financial, and operational features that will enhance a cooperative' s chances for success, all else considered.

We could, in fact, formalize these ideas in terms of an equation with the probability of success, a variable ranging from zero to one, being explained by the economic, organizational, financial, and opera-tional factors. undertake this task, a first point to note is that we do not actually observe the underlying probabilities of success or failure in an enterprise, but, rather, we observe the outcome: a successful business or one that is not successful.

The goal of this section is, therefore, to take the information on successor lack thereof as provided by the respondents in our sample of new cooperatives and statistically relate the successful/not-successful variable to the other economic, organizational, finan-cial, and operational information provided in the surveys. In this manner, wehope to obtain equations that indicate the importance of the various factors in determining success and can be used for predicting the probability of success for a cooperative with a particular set of characteristics.

The statistical technique used to accomplish this task is ca!Jcd /ogit analysis.u The first task is to con-struct a measure of success. Although respondents were given four success/failure categories, these must be condensed into two for present purposes. The choice made was to employ the following dichot-omy: major success and not a major success. Coop-eratives rated as a major success by the respondent were placed in the major success (MS) category and assigned for statistical purposes the value 1.0. All

~For those who have some familiari.ty with statistical meth<>ds, logit analys1S is a modification of the basic linear regression model to acrommodale qualltabve dependent variables such as buy-no buy decisions, yes-no responses, or, as in our situation, the dichotomy between su~cess and lack of success. The technique is discussed in most modem econometrics textbooks.

41

other responses, minor success, too early to tell, and not successful, were lumped into the category "not a major success," (NMS) and assigned the value ofO.O.

At the outset we should note that there is some arbitrariness to this categorization. In particular, each cooperative' scategory is based on theopinionof the respondent. Others might view the matter differ-ently. In addition, placing too-early-to-tell coopera-tives in the NMS may miscategorizc cooperatives ultimately destined for success.

Two factors mitigate the first problem. Multiple survey responses from the same cooperative always resulted in the same success rating.. and statistical analysis of the ratings indicated no persistent bias in the response due to the position of the respondent. In particular, manager/employee respondents were not significantly more likely to rate a venture as a major success than were member/director respon-dents.

As to concern over miscategorizing too-early-to-tell cooperatives, evidence from the surveys tended to suggest with only a few possible exceptions that this was, in fact, an equivocal response. In other words, people choosing the too-early-to-tell category could not judge the cooperative to be a major success but believed there was still a chance to "tum the comer" to success. As such, most of the too-early-to-tell cooperatives would be appropriately placed in the not-a-major-success category.

2. Variable selection.

The number of possible detemrinants of success included in the survey forms exceeds the number that would be prudent to include, given our number of sample cooperatives, in a single equation. Our solution to this problem was to adopt a two-stage approach to the estimation. In stage I we classified our detemrinants of success according to whether they were economic, organizational, financial, or op-erations/management variables. Separate equations were estimated to relate the MS or NMS success variable to each group of detenninants. The most significant determinants from the Stage I equations were used to produce a ''best" model in Stage II.

Thus, we have four Stage I models. Model I relates the probability of MS to several of the eco-nomic factors discussed in Part 1 and summarized in Table 7. For statistical purposes, the three response categories, A, B, and C, needed to be condensed into two: thus for the following factors we created an indicator variable which is set equal to 1.0 for respon-dents who listed A. an important factor and is set equal to 0.0 for all other responses:

Q7a - Price too law (high), Q7b- Existing services ineffective, Q7c - Variable and uncertain prices, and Q7d - Undependable/nonexistent market outlet.

One other economic variable included in Model 1 was an indicator variable to discern if the type of cooperative in our sample affected the probability of success. We set the variable, main products marketed equal to 1.0 if the cooperative marketed fruits, vege-tables, or nuts and equal to 0.0 for all other coopera-tives, e.g., supply and service co-ops orother market-ing co-ops.

Model 2 includes several of the membership factors discussed in Part 2, Section IA. The specific variables are the growth coeffident, which measures growth in membership from the initial planning stage to start up;open membership, an indicator vari-able set to 1.0 for co-ops with an open membership policy, set to 0.0 otherwise; nonmember business, an in-dicator variable set to 1.0 if the co-op accepted non-member business, set to 0.0 otherwise; and the num­ber ofmembers involved at the initial planning stage.

Model 3 examines the role of initial financing factors on the probability of major success. All vari-ables in model 3 were indicator variables and in-cluded: commercial bank loan, set to 1.0 for co-ops with a corrunercial bank loan, set to 0.0 otherwise; Bankfor Cooperatives loan, set to 1.0 for cooperatives with a loan from a BC, set to 0.0 otherwise; have equity redemption plan, set to 1.0 for co-ops with an equity redemption plan, set to 0.0 otherwise; and obtained funding from grant(s), set to 1.0 for cooperatives that obtained one or more grants, set to 0.0 otherwise.

Model 4 covers the role of operations factors including management in affecting the MS or NMS outcome. All variables in model 4 were indicator variables and included the following: feasibility study, set to 1.0 for co-ops that reported conducting a feasibility study, set to 0.0 otherwise; full-time pro-fessional management, set to 1.0 for cooperatives with ful/-limeprofessiona/ management, set to 0.0 other-wise; manager respondent, set to 1.0 if the respondent was part of professional management, set to 0.0 oth-erwise; private consu//ant(s), set to 1.0 for co-ops that engaged a private consultant, set to 0.0 otherwise; and public consultant(s), set to 1.0 for co-ops that used one or more public consultants, set to O.Ootherwise.

3. Results.

Results from estimating the four Stage I models are reported in Table 10. To interpret the results note that the estimated coefficient measures each factor's effect on the probability of a major success based on the data. Therefore, factors with a positive coefficient

42

Table 10. Stage I Logit Analysis of Success Probability estimated absolute

Model/Variable coefficient t-statistic

1. Economic factors Price too low (high) 0.036 0.065 Existing services ineffective -0.339 0.642 Variable and uncertain prices 0.173 0.304 Undependable/nonexistent outlet -0.377 0.687 Main products marketed --0306 0.564 Constant 0.312 0.474 Likelihood ratio test 1.275

2. Membership factors Growth coefficient -1.015 1.633 Open membership 1.025 1.466 Nonmember business 1.286 IBW Number of members at initial planning stage 0.053 2.446"

Constant -2.110 2.149" Likelihood ratio test 12.445"

3. Financing factors Commercial bank loans -0.333 0.498 Bank for Cooperatives loan 0.239 0372 Have equity redemption plan 1.271 2.049" Obtained funding from grant(s) 0.338 0.501 Constant -1.058 2.009' Likelihood ratio test 6.5%

4. Operations/management factors Feasibility study -1.332 1.993" Full-time professional

management 2.404 3.183" Manager respondent 0.159 0.235 Private consultant(s) -0.469 0.752 Public consultant(s) 0.210 0.282 Constant -0.990 1.231 Likelihood ratio test 17.762"

"Indicates statistical significance at the 90 percent level

are associated with an increasing probability of a MS, while factors with a negative coefficient are associ-ated with a decreasing probability of a MS.

However, the coefficients reported in Table 10 are only estimates that may well deviate from the "true" value. The "absolute I-statistic" column in the table provides infonnalion on the amount of confi-dence to be placed in any estimated coefficient. Jn general, little confidence can be placed on estimated coefficients that have a small (close to zero) I-statistic. Confidence in the estimate increases for larger t-statistics. One rule of thumb is that for I-values

greater than 1.65, we can safely assume the troe effect is not zero with 90percentconfidence. All I-statistics that meet this cutoff are noted with an asterisk in the table.

Another feature of Table IO to note is the likeli-hood ratio test statistic reported for each model. This statistic measures the statistical significance of the overall model, not any one particular variable in the model. Once again, the larger the value of the statistic the more confident we can be that there is some significant power in the model to explain the MS, NMS dichotomy. Values of the statistic that meet the

43

90 percent confidence level cutoff arc noted with an asterisk.o12

Turning to the individual models, Model 1, eco-nomic factors, perfonned the poorest. None of the estimated coefficients met the 90 percent confidence level cut off nordid theoverall model as measured by the likelihood ratio test. These results do rwl mean that the economic factors cited in Part 1are notcrucial to success. Rather, the results probably indicate that the economic conditions are all of nearly equal im-portance and are thus unable to help explain the dichotomy between MS and NMS.

Model 2, membership factors, did do a good job of explaining success. The overall model meets the test of statistical significance as do most of the indi-vidual coefficients. All have the signs we would expect based on the analysis in Part 2, Section IA. Open membership and using nonmember business are both positively associated with MS. This result probably reflects the overriding importance for new cooperatives of generating as large a business vol-ume as possible to exploit the available economies of size in production and marketing.

Model 2 also indicates that the greater the num-ber of members involved in the initial planning stages, the more likely is success. More involvement means that the organizing costs are spread across a greater number ofpeople, so no one bears too large a burden. The sign on the growth coefficient (GC) reflects the importance of expanding the member-ship beyond the initial coalition prior to start up--smaller values for CC are associated. with larger rates of membership growth. Thus, the negative sign on GC means that the greater the growth rate, the more likely is success.

Model 3, financing factors, overall did not quite meet the 90 percent significance level cut off but contains some illuminating results, nonetheless. Debt financing from a commercial bank was nega-tively associated with the success probability while BankforCooperativcsfunding was positively associ-ated with success. Although neither coefficient meets the 90 percent significance test, the results do provide some evidence to support the importance to new co-ops of exploring the advantages of BC fund-ing. It should be noted, however, that the positive correlation between BC funding and success may result from the BCs willingness to loan to only the potentially most successful cooperatives.

Also positively correlated with success, though not statistically significantly, was the variable indi-cating receipt ofone or more external grants. Finally,

presence of an equity redemption plan was posi-tively and significantly correlated with success. This result probably reflects both the good member rela-tions aspcctsofhavinga visible plan in place and the fact that the stronger cooperatives were most likely to have a plan.

Model 4, operations/management factors, was also a quite successful model, easily meeting the 90 percent cut off for overall statistical significance. The key variable in the model is the one indicating pres-ence of full-time professional management. It is positively correlated. with success and is h'ghly sig-nificant.

Also interesting to note is that use of public consultants was positively correlated with success while use of private consultants was negatively cor-related with success. Although neither coefficient met the statistical significance test, they do provide some evidence on the efficacy of taking advantage of the publicly available expertise.

We note that manager respondents were some-what more likely to judge a new co-op to be a MS than were member /directors. However, both the coeffi-cient and its I-value arc very small, prompting our earlier conclusion that position of the respondent did not significantly bias the surveys.

Finally, the negative sign on the variable indicat-ing thata feasibility study wasconducted is an anom-aly. Oearly conducting a feasibility study should not diminish success prospects. It may be that feasibility studies were commissioned. only in those cases where success prospects were most dubious, thus ultimately leading to the negative correlation be­tween MS and a feasibility study.

Table 11 contains the estimation results for the Stage II "best" explanatory model of the MS, NMS dichotomy. Factors chosen from the Stage l models were the following membership factors: growth coefficient, number of members at the initial plan-ning stage, and use of nonmember business. Also included were the variables indicating presence ofan equity redemption plan and full-time professional management. The respondent's position variable was also included to control for any biases caused by that factor.

As Table 11 indicates, the Stage II model performs quite well, easily meeting the standard for overall significance. All individual factors have the expected signs with the variables indicating presence of an equity redemption plan and full-time professional management meeting the statistical significance cut off and the variable indicating the

"'The likelihood ratio statistic ls distributed accorctmg to the chi squared distribution with degrees of freedom equal to the number of estimated parameter•.

Table 11. Stage II Logit Analysis of Success Probability

Estimated Absolute Variable coefficient t-statl.stic .

Growth coefficient Number of members at initial planning stage Nonmember business accepted Haveequity redemption plan Have full-time professional management Respondent's position in co-op Constant Likelihood ratio test

-0.573 0.035 0.756 1.163 1.256 0.259

-2580

0'84 1.561 0.991 1.695• 1.773• 0389 2.665

19.197-

•Indicates statistical significance at the 90 percent level.

number of members at the initial planning stage only narrowly missing the cut off. The "R2" statistic indicates that this model explains about 30 percent of the variation between MS and NMS found in our sample, a good pcrfonnance by legit model standards.

V. Summary. Titis part of the report has been devoted to de-

scribing our survey of recently-formed American agricultural cooperatives and to analyzing the sur-vey responses that were generated. We found, not surprisingly, a mixed bagof success (about halfofour sixty plus respondents reported that their co-op was a major success) and other, less serendipitous results. Respondents generally agreed that the economic conditions amenable to co-op formation discussed in Part 1 were, in fact, important factors motivating the decision to develop a cooperative.

Most of our sample cooperatives had open membership and accepted nonmember business. These characteristics we found to be positively corre-lated with the success probability based on the statis-tical analysis reported in Section IV. A relatively few people were found to be involved in the initial organ-izing stages of the cooperatives, but the more that

were involved, the more likely was the co-op to be successful. Similarly the statistical analysis demon-strated that membership growth from the initial planning stages to the time of start up was also important to success.

On the topic of financing about two-thirds of the sample cooperatives generated initial equity capital from membership fees, and 43 percent issued stock. Only aboutone-fourthobtained start-up capital from grants.

About one-third of the sample cooperatives ob-tained loans from a Bank for Cooperatives, and one-fourth borrowed from oommerdal banks. The BC borrowers tended to be the more successful of the

·=· In the area ofdecision making and management, the new cooperatives made extensive use of both public- and private-sector consultants, with co-ops using public sector assistance tending to be more successful. Voting with two exceptions wasbased on the one person, one vote principle. Finally, a key success-detennining factor was found tQ be the pres-ence of full-time professional management which 59 percent of the sample cooperatives indicated having.

45

CONCLUSIONS AND RECOMMENDATIONS

Cooperatives have been an important part of Step6: Finance initial capital outlays, and America's agricultural economy, and they no doubt will continue to play a major role as agriculture

generate a sufficient equity base by using flexible membership fees

prepares to face the demands and challenges of the and possibly gTants. Obtain debt twenty first century. However, to retain their vitality capital through the Banks for cooperatives must be flexible and in step with the Cooperatives or Industrial Devel-modern, evolving economy. opment Bonds.

In focusing on the role that cooperatives can play Step 7: Develop a plan to refund retained in the modern economy this report has confinncd equities back to members. some traditional maxims for cooperatives' behavior Step 8: Establish pricing policies that but has also presented some ideas at odds with exploit cooperatives' built-in traditional wisdom. We state our main conclusions flexibility in pricing. If possible, and recommendations in ten steps to success to be approximate the optimal prices to followed more or less in sequence. members using the procedures set

Step 1: Analyze market conditions with a keen understanding of what co-operatives can and cannot do. Do not waste time and money on organizing a cooperative when markets already perfonn reasona-bly well.

Step 9: forth in Part 2, Section IIIA. Carefully consider membership policies. There are prospective advantages and disadvantages to open versus closed membership and to accepting or refusing to accept nonmember business. These considerations are dis-

Step 2: If a determination is made based on the criteria described in Part 1

cussed inPart2,SectionIA. Deter-mine which considerations are

Step 3:

of this report that a cooperative can potentially generate net bene-fits to fanners, conduct a feasibil-ity study questionning whether sufficient membership, business volume, and equity capital can be obtained to realize these benefits. If Step 2 generates an affirmative response, organize the coopera-tive so as to maximize the mem-bership size and to build commit-ment among members. Beflexible as to financial commitments and voting procedures to achieve the first objective. Use long-term membercontracts with stiff penal-

Step 10:

dominant for each particular situ-ation. Always remember the relative strengths and weaknesses of co-operatives as a business organiza-tion. The strengths include har-monization of trade, ease of com-munication, pricing flexibility, and government policies that are beneficial to cooperatives. The main weaknesses of cooperatives are their difficulty in obtaining equity capital and their failure to reward entrepreneurial activity. Cooperatives may also be less flexible than other business or-

Step 4:

ties for violations to achieve the latter. Carefully estimate the new cooperative's business volume and plan capital facilities that will efficiently handle that volume.

ganizations owing to their demo-cratic nature. Exploit the strengths and take steps to over-come the weaknesses. Ourhopeis that this report helps provide the means to do so.

Step 5: If possible, hire a full-time professional manager to run the operation.

46

• • • •

• •

• •

• •

•• • • • •

• •

•• • •

• •

i ' •j '! • ! ' 1•' '• ;

!•' ' ! !

'

' ' ' " l

'• ' '

'

•

• •• ••

•

. . . ." ,, "~

•

.~i ' ~~

=~-.] ;

!' -:;

= -·••-..• .•• !: :-• .•=• ~~

.::• -:;.•

.• ~ •. ~

·-

• •

'•• •

' '• ' ' '•'' '

•

• •

i•

' 1 ;

' ' ••

' l

' •' '

• • • •

• •

: : '

l'• ' •

j' ' i ii. •

' ! ;' l

' ' f ' '

'

•l

' ' i .•

'; '

'j l ' f '

' • ' ••; ' ' ' -

•• •' '' ' •; • • ' '! • i '

! '' i •' ' ! !J

! ' ' ' • ' i ' i ! ! ' ! •i • ' ; ' ' ' ! l ! •• '' l !

47

l

• • •

• •

• • ••

• •

• •

•• •••

• •

• • •

• •

• • ••

•

• •

•••

• •

• •

• • •

•••

••

•

• • •

•

'.• ' •j'.... .~ • ..,. ....•• • .. •

. i;

• '~ ' ;~ •..• ...• ...• ••

'ti ~. ~ ... ... • '• ~.;;:' . . g . !;....g ,,·'•• •' • ~ .;. i.• . ' .• •. •' ....• ••• ' ~ ~ ~lff •' 'll .c "• .......: = ' l ' i! -..~ g~,:;!~ •..' • • .!.....l •. .• '• ·-• .•••-·-.

•!'" l:;•

....'.= . ' ~ .~ . "! i >~ " '• •• .l I

'• •

...... ~~ .. !i.... l• i! .l: i ... ~ .••..• • ..••• • . .' '' ••

! •' '• •i • j'' ,. '

' ••• '.,• i ' i .:' 8 l1 '• .•

' ~ ;; . l l . . '•'

• • !• ;! •! ' ' • '•• '• ••• '' ..

; ~ !

•

' .,.••' l ' ....' i •'

•! ' ,ll i ! ! ] ~€~ •.. l•1

•• • • ~::. • . i ' •

• ..'·1 .,_ ij-::;• ' ' : : ~ ....

!

'•' ' • 1§

• • F..l: ... w

1 1" l• ~ .... • i .. ,.::::: ••; ,· ... ! .,• _,...., 0i !....! ~ l 'l •• • l ••• ..

•

48

••

•• • ••• •••

••• •••

••

• •

••• • • •••

•• • • •

•••

• • ••

• • •

• ••

••

• •

•• ••

• •

l

•

• ••=. :• • •...... l ~ ..

:." ~ ~ i: ::..•••••••••.. ; : .• •iii ~~ti :: ~ ~ ~ .< • ~ .. :·---.• • • ~i .. l !' •• 0 - g.' .. ..... ~ - t

• <

•

..•• ~t

'':! ;:: ~ ·-..• • ~l <

~~..• •.,

• •

1•

' •

• •

• •

=~~.-. ~~~ • ••

' '·. ' •

"• ..

... " ..

•

:••::

-: i••;::. .

·~

:1•• ..• ~s., ; ~ g... c • :... !!~..• • ~ • "

·~ ~ ,--, ' ' =.='· .• • ' ..•• ::~

t: .. ! .. ... •••..• 0 ~

. ''.'.. ''l

.: !••g

l.: ;: ~ . '.. .• •..• ••

'·!.. ••.. :;::.•• ,· !~ •

0 "

;

'• ; • }

,, • "

' l'

t !•

•l

,. e 8~ ': •! i~• ~ ~

~',: !

•;

' ~~ !: !••.. ~ ;' .I ! ' l

' l ' • 't I j • : •

j ' f ' ' !

11 I I..

•

' !••j

' •' i

'

•·-;;;•' . •

...' ·- ·-1= !: ;,l.• :•!! ·-•.' ' • ..••. ~~ I: ~ ::s ! •.. ... •

• '

''

l I

I• i '• ! ~ j

'•

• ' ;' l •"

•

! ! :•

t ! ! '

' I ! •"

49

• • •

•• • •

• •

=i'

~ ' '

_,• : ;--••

•' ' • •' •

~ •

• • < ' -· '

!

•

•

' •-•• ..• '••' -~= . '•• ' ,••. •.. ='ii . ~ -;

' '

'• • ' •• .. • .' ,, l'

' •

•

;• •• ' •'•l • ' ' •• ' ' '• •'

'• ••f ' ' j!i

j ~ ' •'•5 j' ' i ' ' • ]•

l ' • '

g

' !' '! . i

j '

.! ! i '' l ; i ' !• • J '• ' •

• ' •' i ' i ~ • ' j l ' •

• ' ' •

i' '! !' }l ' ' ! ' l '• ~ • j '! .. i i. •''' . l 'l

; ' ' ' ! 1 ! J i

" " • ••' so

•••

• • •• •• •

•••

• • • ••

••

••• •• •••

••

•• ••

••

••

• •

••

•

l ' '• 1 •' 1 ' •• i I

•

.. <l <>... ~ 0 ..

•. ;;... ••'. e•.- I~~...·- :• :: . ··--. =ti !!E.. ..,. ~ei c: ~..-... r~

.-.-..··- -a :: ~ •

••

' •

' •l•

' ' ' •

'.il ,.'' .-·'......·- ·: ,,=:

•

' •l l•'

!•

•

' ' f• ' ' ..••• .' I!

•

•

. ' : .::; .!:. •8:: ii ' ,, ~ !., . • ' . ~ f.] . ' '1 ~ .' t

:] .-. .• ~ = .:-~ '' - .' '

·111111111

•

1 •'

•l •

l ! l

'•i ' '

!

j

' ' '

!

' i •' ' j l

•

i• '• • •. .:.: e' ' ~~

•~ t . ••' -.. ••••• -..-..-.. ' . 1~ .-j~

•

! ' ' ' ~ •

' ' ' • ! 1• ~ l

• •

i' '•' i

'•

i •

i •

i~• ..•.. -··-• i"il..' I'..

-." •••

-. •

..·-·-• • 0 .. -:;

••~ .s g.-. ... :~ !; ~ :

!]~ .•• .... ! :", i.

• •

•

• ' ' l ' '

'

'• ' ' 1 l

;• ';;-• .• ...-: ' 'I ;,, • '

51

••• • • •

• •

• •

••

•• ••

• •• • • ••• • • •

• • •

• ••

• •

• • ••

••

• ••

l

i i•

•l ~

!•' 1

i !

• '• I

I If, :.::'.1 ~ 8: ••• -. .s ,.1;;:.

! :: ':;

l

! :

! •

I •"

11 ll ..' if. 11 ··-! l

. I !"•1• I I! ..• I a~: .. ~~•••.. ~-·~•~ •....• "l "Q'll l at 11.::

1 l••

l• ' •

• I

' :• i

••

'•<I

! l ': !• l ~ •I l j

•'

'l ! ! 5

' t

•I I

.'•

.• ~! I:

:]

·: ••• ..••• •

l: .. !!; : .. ~ !;

~ i= !' '!...ii :.x"! ... .; .!l ;: ~ ~ = ; : ! .:; • u ...•• : ~I ~ ==

52

~~ ' :: ~ ';::'•• .•• ,,·-•' · l• ~ ~

.• ' !g: .... ~ ~

". !i g~ ., l [...' ...•••.. a~:

:~~.. :... . ~ .. '

I n aoco<Oanco """ OWi "'Ole S'°'" ano F_., IOw> on 0 Un"o""ty pol 1oy, tho U n "'""' of C"•'o m " doos n ot "'"'" m•noto 1n ""' ol 1to pol<CIOO. o<O<OO "r••- or pract1c ..on oho """'Of ""'·color, nao •On•I o" ' 1 n, reJ•o•on , ••• , '"'""' onomo,.on , honO'cap, ago, vo•o<ans ""' "'· m"'"'' cono ""'" (•• <IOflnod on S..CO '"" ' »26 of\n o C."00 rn 10 G"'"'"""'"' Cod o ). anc0>t')', or mor;ta,.tat" >: no< Ooo• '"" Un.,or>lty dl><nminaoo "" Che "'"" ol '""""""'"· """'" Oho "m11> 1m•<»Od o, law o< un..or>;oy policy In conlo<m.once wdh appl<cab!o law and Um"'"'"' pol>Cy, tho Un•..,,lty o1C""•'"" " an afformotlVO ac.,on/OQUOI oownunoty omp10,o< lnQum•• ""'""''"" t"8 Uni"'"'"• "'"'' oooort"nlt, pollc10• moy ba d•re<!OO to tho v;co Choncollo< o1 AeoOomlC Aff•,,•-Aff1""°'"' Aco;on Off"'' and T"" IX Cooro1ne•or, ~1 """" "'"· (91e1 75•·2<170 Spo<>eh '"" hoar1no ompo1<eO """""' m•y OIOI 7~-132<1 iTOO),